The Jewish Context of Jesus’ Miracles

The Jewish Context of
Jesus’ Miracles

Eric Eve

Journal for the Study of the New Testament

Supplement Series 231


Copyright © 2002 Sheffield Academic Press

A Continuum imprint

Published by Sheffield Academic Press Ltd

The Tower Building, 11 York Road, London SE1 7NX

370 Lexington Avenue, New York, NY 10017-6550

All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording or any information storage or retrieval system, without permission in writing from the publishers.

British Library Cataloguing-in-Publication Data

A catalogue record for this book is available from the British Library

ISBN 1-84127-315-5

This book is dedicated to my mother,
Barbara Eve

And to the memory of my father,
Jack Eve




List of Abbreviations

Chapter 1

Definitions and Scope

Literature Survey


Chapter 2
Miracle in Josephus


The Concept and Language of the Miraculous in Josephus

The Distribution of Miracles in Josephus

The Function of the Miraculous in Josephus


Chapter 3
Miracle in Philo


The Concept and Language of the Miraculous in Philo

The Distribution of Miracles in Philo

The Function of the Miraculous in Philo


Chapter 4
Miracle and Wisdom


The Wisdom of Solomon

Ben Sira


Chapter 5
Miracle in Pseudo-Philo


The Concept and Language of the Miraculous in Pseudo-Philo

The Distribution of Miracles in Pseudo-Philo

The Function of the Miraculous in Pseudo-Philo


Chapter 6
Enochic Literature


The Book of Watchers in 1 Enoch


Conclusions and Developments

Chapter 7
Selected Qumran Texts


The Genesis Apocryphon (l QapGen)

The Prayer of Nabonidus (4 QprNab)

The Messianic Apocalypse (4 Q521)

Some Demonological and Apotropaic Texts


Chapter 8
Miracle and Romance





Chapter 9
Miracle in Second Temple Literature

General Observations

Agents of the Miraculous

Functions of the Miraculous


Chapter 10
Charismatic Holy Men


Ḥoni the Circle-Drawer

Ḥanina ben Dosa


Chapter 11
The Sign Prophets


The Accounts of the Sign Prophets

Characteristics of the Sign Prophets

Chapter 12
Jewish Exorcists


New Testament Evidence

Jewish Evidence

Later Non-Jewish Evidence


Chapter 13
Healers, Magicians and Spirits


Healers (Medical Anthropology)


Spirit Possession

Chapter 14

General Observations

Healing and Exorcism

Other Miracles

Jesus as Prophet—and More than a Prophet


Index of References

Index of Authors


The present study has grown out of my doctoral thesis. In the thesis I somewhat ambitiously aimed to survey most of the main Second Temple literature for evidence of Jewish views on miracles and miracle-workers. The limits on thesis length (and the requirement to finish the thesis within a reasonable length of time) inevitably resulted in many of the texts receiving only the skimpiest treatment. In the present study I have attempted to correct that by adding a further four and half chapters. This has still left a number of texts to be discussed, or merely referred to, in summary fashion in the chapter on ‘Miracle in Second Temple Literature’, but at least the range of texts that have been discussed in some depth has now been substantially widened. Moreover, I suspect that few of the remaining texts would have yielded results commensurate with the space needed to discuss them in depth. The major omissions probably lie elsewhere: ideally a study of this sort should also have compared the Septuagint with the Hebrew Bible, and examined the Targumim and Midrashim on the assumption that they may well contain traditions going back to Second Temple times. But to have attempted to take on these tasks as well would have expanded this study beyond manageable bounds.

This study would not have been written at all but for the help and support of several institutions and individuals to whom my thanks are due. First of all I should like to thank the Principal and Fellows of Harris Manchester College, Oxford, for providing me with the most pleasant and welcoming of academic homes, first as a student and now as a Junior Research Fellow. I should also like to thank Middlebury College, Vermont for funding my Junior Research Fellowship, and the Humanities Research Board of the British Academy for funding my doctoral research prior to that.

Several individuals have contributed to this study in one way or another. My examiners, David Wenham and Loren Stuckenbruck, both made a number of comments on my doctoral thesis, and their detailed criticisms have proved a valuable spur to carrying out the additional work included in this published version. I am particularly grateful to Loren Stuckenbruck for drawing my attention to a number of additional texts and sources that needed to be taken into account. Martin Goodman also merits my gratitude on a number of counts: more generally, for helping to inspire an interest in Second Temple literature when he taught me as an undergraduate; more specifically for supervising my Master of Studies dissertation which now appears, in slightly revised form, as the chapter on Josephus. His advice on how to tackle the works of Josephus has been taken up in subsequent chapters, so his influence extends beyond the one chapter he was closely involved with, and I am also grateful for his comments on an early draft of my Qumran chapter. Above all, though, I should like to thank John Muddiman, my graduate supervisor and undergraduate New Testament teacher, for many years of encouragement, friendship and support, and for his patient reading both of my excessively long undergraduate essays and of several drafts of my doctoral dissertation. It is in no small measure due to his inspirational teaching that I chose to work in the field of the New Testament, and he also contributed many detailed and helpful suggestions to the doctoral thesis that forms the basis of this study. It is conventional but in this case entirely true to add that while all these people have helped to make this study vastly better than it would otherwise have been, its many faults and shortcomings remain entirely my own doing.

There is, as I have discovered, far more to getting a book into print than simply writing it, and my thanks are due to all those who have assisted one way or another in the publication of this work. In particular, I should like to thank the many people associated with Sheffield Academic Press who have contributed in one way or another, not least the series editor Stanley Porter for recommending publication of my manuscript in the JSNT Supplement series, Heidi Robbins for patiently fielding the barrage of questions I emailed her while preparing the manuscript for publication, and Heather Haig-Prothero, for steering this book into production and for her unfailing courtesy in replying to my daily queries while I was checking the proofs. In this connexion I should also like to record my special thanks to Christabel Powell for interrupting her work on Pugin to help check the proofs, and for her constant love and companionship not only throughout the time this book was written but also during my time as a theology student beforehand.

Finally, I should like to thank my parents, whose support and encouragement proved an enormous help through my second Oxford career (and, indeed, through my first)—neither they nor I imagined when I graduated from Brasenose with a degree in Engineering Science in 1975 that I should be returning to Oxford 17 years later to read for a BA, MSt and DPhil in theology! Their generosity in providing me with a place to stay and work during university vacations (or when I just wanted a change of scene) contributed substantially not only to the success of my studies, but to making my mid-life change of career financially possible. Regrettably, my father did not live to see my doctorate awarded or this book completed, and so it is to his memory, along with my mother, that this work is dedicated.


AB    Anchor Bible

AGJU    Arbeiten zur Geschichte des antiken Judentums und des Urchristentums

AnBib    Analecta biblica

ANF    Anti-Nicene Fathers

ANRW    Hildegard Temporini and Wolfgang Haase (eds.), Aufstieg und Niedergang der römischen Welt: Geschichte und Kultur Roms im Spiegel der neueren Forschung (Berlin: W. de Gruyter, 1972–)

BAGD    Walter Bauer, William F. Arndt, F. William Gingrich and Frederick W. Danker, A Greek—English Lexicon of the New Testament and Other Early Christian Literature (Chicago: University of Chicago Press, 2nd edn, 1958)

BARev    Biblical Archaeology Review

BETL    Bibliotheca ephemeridum theologicarum lovaniensium

Bib    Biblica

BibOr    Biblica et orientalia

BJS    Brown Judaic Studies

BNP    Bearer of Numinous Power

BNTC    Black’s New Testament Commentaries

BTB    Biblical Theology Bulletin

BZAW    Beihefte zur ZAW

CBC    Cambridge Bible Commentary

CBQ    Catholic Biblical Quarterly

CBQMS    Catholic Biblical Quarterly, Monograph Series

CSCO    Corpus scriptorum christianorum orientalium

DJD    Discoveries in the Judaean Desert

FRLANT    Forschungen zur Religion und Literatur des Alten und Neuen Testaments

GAP    Guides to the Apocrypha and Pseudepigrapha

HSM    Harvard Semitic Monographs

HTR    Harvard Theological Review

HUCA    Hebrew Union College Annual

HUT    Hermeneutische Untersuchungen zur Theologie

ICC    International Critical Commentary

IEJ    Israel Exploration Journal

JAAR    Journal of the American Academy of Religion

JBL    Journal of Biblical Literature

JBLMS    Journal of Biblical Literature, Monograph Series

JJS    Journal of Jewish Studies

JRAS    Journal of the Royal Asiatic Society

JRS    Journal of Roman Studies

JSJ    Journal for the Study of Judaism in the Persian, Hellenistic and Roman Period

JSJSup    Supplements to the Journal for the Study of Judaism

JSNTSup    Journal for the Study of the New Testament, Supplement Series

JSP    Journal for the Study of the Pseudepigrapha

JSPSup    Journal for the Study of the Pseudepigrapha, Supplement Series

JTS    Journal of Theological Studies

LCL    Loeb Classical Library

LSJ    H.G. Liddell, Robert Scott and H. Stuart Jones, Greek-English Lexicon (Oxford: Clarendon Press, 9th edn, 1968)

MNP    Mediator of Numinous Power

NCB    New Century Bible

neb    New English Bible

NIGTC    The New International Greek Testament Commentary

NovT    Novum Testamentum

NovTSup    Novum Testamentum, Supplements

NTS    New Testament Studies

OTM    Old Testament Message

OTP    James Charlesworth (ed.), Old Testament Pseudepigrapha

PNP    Petitioner of Numinous Power

RB    Revue biblique

RevQ    Revue de Qumran

RGRW    Religion in the Graeco-Roman World

rsv    Revised Standard Version

SBL    Society of Biblical Literature

SBLDS    SBL Dissertation Series

SBLSP    SBL Seminar Papers

SBT    Studies in Biblical Theology

SC    Sources chrétiennes

SJLA    Studies in Judaism in Late Antiquity

SNTW    Studies of the New Testament and its World

SPB    Studia postbiblica

STDJ    Studies on the Texts of the Desert of Judah

SVTP    Studia in Veteris Testamenti pseudepigrapha

TSAJ    Texte und Studien zum antiken Judentum

TU    Texte und Untersuchungen

VTSup    Vetus Testamentum, Supplements

WUNT    Wissenschaftliche Untersuchungen zum Neuen Testament

ZAW    Zeitschrift für die alttestamentliche Wissenschaft

Chapter 1


Definitions and Scope

In relation to the Bible, ‘miracle’ is a potentially misleading term. For people in the modern West, it is coloured by the eighteenth-century debate on whether ‘miracles’, understood as God’s infractions of the laws of nature, are at all possible. This in turn presupposes a worldview only made possible by the Newtonian revolution in natural science. A handful of philosophically sophisticated writers in antiquity also believed that nature was subject to immutable laws that made miracles impossible, but the popularizing of this belief only came about as a result of the Enlightenment. No biblical writer shows awareness of such a view.

Yet the word ‘miracle’ has traditionally been used, and continues to be used, to describe many extraordinary events narrated in the Bible, including the exorcisms, healings, and other mighty works of Jesus, deeds for which the Evangelists use words such as δύναμις or σημεῖον. If we are to continue to use the word ‘miracle’ in this context it cannot bear the meaning it acquired in the Enlightenment. A biblical ‘miracle’ is not ‘a transgression of a law of nature by a particular volition of the Deity, or by the interposition of some invisible agent’. It is rather a strikingly surprising event, beyond what is regarded as humanly possible, in which God is believed to act, either directly or through an intermediary.3 All creation is thought to be under God’s control; a miracle occurs when God chooses to exercise that control in an unusual fashion.

Yet many of the miracles narrated in the Bible also appear to be ‘transgressions of the laws of nature’. The multiplication of loaves and fishes, for example, would violate either the Law of Conservation of Matter/Energy (if we suppose the additional food to have been created ex nihilo) or the Second Law of Thermodynamics (if we suppose it to have been created through the ordering of existing matter). Such laws are so fundamental to modern science that they go beyond being merely well-tested theories. If we are to reserve ‘miracle’ to denote a biblical concept, we need another word to describe a supposed exception to the laws of nature; in this study, the word ‘anomaly’ will be used for this purpose. Anything violating so fundamental a principle as the Second Law of Thermodynamics would be a ‘hard’ anomaly indeed; anything for which it was merely difficult to devise a plausible scientific explanation would rate as a ‘soft’ anomaly. Whether or not either type of anomaly can actually occur will not be debated here. The position taken is that while neither type of anomaly can strictly be ruled out a priori, the quality of evidence required to establish the occurrence of an anomaly increases with its ‘hardness’. On the basis of strict historical enquiry, in no case is the biblical evidence sufficient to establish the occurrence of a hard anomaly (such as multiplying loaves and fishes, turning water into wine, or resuscitating a man four days dead in the tomb). Concerning soft anomalies, however, one may be more open-minded.

Many modern scholarly discussions of the kind of phenomena this study is concerned with tend to use the term ‘magic’ in connexion with them. This study, however, will preserve a distinction between ‘magic’ and ‘miracle’. Whereas a miracle is a strikingly surprising event that is thought to be an act of God, magic is a deviant procedure in which some feat regarded as beyond normal human ability is brought about by some agency other than God. It must be stressed that this distinction is an emic rather than an etic one, that is that there may be no objective means for a modern scholar to decide whether something should in fact be classified as ‘magic’ or ‘miracle’ in this sense. But that is not the point at issue; the point at issue is how people in the first century might have classified things. This allows that even in the first century people might have disagreed, and disagreed sharply, over whether something should count as ‘miracle’ or ‘magic’.

The use of the term ‘deviant procedure’ clearly invites the question, ‘deviant from whose norm?’ It is also meant to highlight the fact that behind the terms ‘magic’ and ‘miracle’ lie issues of legitimation. In trying to understand first-century views on miracle, these issues are of sufficient interest that the proposed distinction between ‘magic’ and ‘miracle’ seems worth preserving.

One might have expected a study involving miracles, Jesus and Judaism to be called either ‘Jesus’ miracles in their Jewish Context’ or ‘Late Second Temple Jewish Views on Miracle’. The reason for avoiding the first title, attractive though it may be, is that it would entail too much repetition of the discussion of Jesus’ miracles in recent literature while not leaving enough space for an exploration of their Jewish context, which is precisely the area to which insufficient attention has been given (as will be argued in the literature survey). The reason for avoiding the second is that it is nevertheless Jesus’ miracles that this study is primarily intended to illuminate. In the course of doing so it will encompass an extended survey of miracle in Second Temple Judaism, and will accordingly contribute to this area of Jewish studies as well. Nonetheless, the survey will be carried out with the aim of illuminating the significance of Jesus’ miracles in their Jewish context. By ‘Jesus’ miracles’ is meant here both the miracles attributed to him in the literary creations of the Evangelists and the healings and exorcisms of the historical Jesus, though the focus will shift from the former to the latter as this study progresses.

The focus on the Jewish context of Jesus’ miracles is not intended to imply that an understanding of pagan miracle-working is irrelevant. In the first instance it is simply necessary to restrict this study to a manageable topic. But, more profoundly, the study will suggest both how Jesus’ miracles are distinctive against their Jewish background and how their Jewish background nevertheless provides the proper context for their interpretation. To be sure, much of the Jewish literature to be examined has been influenced by Graeco-Roman culture, but it will be beyond the scope of this study to investigate these influences as a separate topic.

Literature Survey

A thorough survey of all the relevant literature would constitute a monograph in itself, so here only a representative sample will be briefly surveyed. There is no need to devote space to the older works on Jesus’ miracles, such as those by Alan Richardson, James Kallas and R.H. Fuller. Neither need the slightly more recent books on Jesus’ miracles that seem mainly apologetic in intent be discussed.6 In the mainstream scholarship of the last 25 years two trends in particular may be noted. The first is an increasing concern to relate Jesus to the Judaism of his day. The second is an increasing inclination to take the miracles of Jesus seriously rather than regarding them as a faintly embarrassing irrelevance, or merely as a later layer of hellenizing accretion. In this connexion books dealing with the historical Jesus rather than specifically with his miracles have become notably important, though special mention should also be made of a few works in the latter category.

The first of these is Graham Twelftree’s book on Jesus’ exorcisms. Twelftree is particularly concerned to set Jesus’ exorcisms against the background of other exorcistic activity at the time, and we shall be thus be engaging with him in Chapter 12.

Whatever its faults, Morton Smith’s Jesus the Magician at least drew attention to the importance of Jesus’ miraculous activity at a time when most mainstream scholarship still preferred to view Jesus almost purely as a teacher. One may accept Smith’s point that Jesus’ miracles contributed greatly to his popular following without subscribing to the rest of his thesis. In the main, however, Smith’s argument that Jesus should properly be understood as a magician relies too heavily on privileging the perspective of his opponents and on discovering forced parallels between Jesus’ actions and those of magicians regardless of context. Although John Hull’s book on Hellenistic Magic and the Synoptic Tradition is less extreme in its claims, it also sets up the problem in such a way that it becomes very hard for anything to count as ‘miracle’ as opposed to ‘magic’. Both books have accordingly come in for a considerable amount of justifiable criticism.10 The question of magic and miracle will be taken up in Chapter 13.

Another book that seeks to shift the focus from Jesus’ teaching to his miracles is Stevan L. Davies’ Jesus the Healer. Davies makes extensive use of anthropological research into spirit possession to make this the key concept for understanding Jesus’ ministry. He unfortunately attempts to press the use of this concept beyond the point where it remains plausible. His book nevertheless contains some fascinating and provocative material, and again will be discussed further in Chapter 13.

Also relevant to Chapter 13 is the third main section, the ‘functional approach’, of Gerd Theissen’s work on the Gospel miracle stories. The first main section of this book, focusing on a synchronic approach, combines form-critical with structuralist insights to produce a detailed analysis of the six basic roles and 33 motifs in the miracle stories and the way in which they interact. This has earned Theissen some stiff criticism for mixing narrative-critical and form-critical approaches from scholars inclining more towards the former.13 Nonetheless, this part of his study remains informative and ground-breaking. It also enables Theissen to make some interesting observations when he turns to his diachronic approach, in which he investigates how such themes as faith as a boundary-crossing motif and the commands to secrecy are transformed through time. Of particular interest is the way in which Theissen’s combination of synchronic and diachronic approaches enables him to question the standard distinction between tradition and redaction, on the grounds that what successive narrators do might better be described as realizing the various potentialities immanent in the motif fields of a given story, rather than as editing a tradition.

The overall view of Jesus’ miracles that emerges from Theissen’s book is that Jesus was unique in employing miracles, which are episodic realizations of salvation for a small number of individuals, in order to transform the world into the Kingdom of God, a transformation that apocalyptic expectation looked forward to as universal. In Theissen’s view the correct interpretation of Jesus’ miracles is therefore eschatological, and this is especially so of the exorcisms. Yet Theissen finds that the eschatological interpretation of miracles largely disappears as one moves from the sayings traditions to the miracle stories. The tendencies of the tradition are to smooth out the most striking characteristics of Jesus and to heighten the miraculous nature of Jesus’ acts to an exceptional degree, thereby popularizing the stories. Essentially the same view re emerges in Theissen’s much more recent textbook on the historical Jesus.

As intimated above, some of the most interesting recent contributions to the question of Jesus’ miracles occur in treatments of the historical Jesus. One may observe the trend towards taking miracles more seriously in books of this kind when one compares Ben Meyer’s four pages (out of 253) devoted to Jesus’ miracles compared with over five hundred pages in the more recent book by John Meier. Scant treatment of the miraculous is also found in books concerned with the historical Jesus (or some relevant aspect thereof) by authors as diverse as John Riches, Martin Hengel, G.R. Beasley-Murray and Hendrikus Boers.16 By no means all these writers are hostile to the idea that Jesus worked miracles; they merely imply by their near silence on the matter that they consider it to be of little significance. Even Tom Wright, who argues vigorously that Jesus’ miracles are to be seen as integral to his ministry, allows himself only 11 out of 662 pages to discuss them.

Wright’s book belongs firmly in what he calls the ‘third quest’ of the historical Jesus, the characteristics of which include the reconstruction of a Jesus who makes sense within the Judaism of his day. The attempt to understand Jesus within the Judaism of his day is hardly a new idea—it goes back at least to Reimarus—but it is clearly an approach that is sympathetic to the concern of the present study. It was given fresh impetus over 20 years before the appearance of Wright’s book by Geza Vermes’s Jesus the Jew. Not only did Vermes place Jesus firmly within Judaism, he also took his miracles seriously enough to attempt to sketch the appropriate Jewish context for understanding their significance. His thesis that Jesus is to be placed among other Galilean charismatic holy men such as Ḥanina ben Dosa and Ḥoni the Circle Drawer has proved highly influential, and is frequently reflected in the later literature (as will become apparent). In Chapter 10 of the present study, however, Vermes’ category of charismatic holy men will be strongly criticized. Vermes’ work nevertheless remains one of the more important milestones along the road to be trodden here.

There has been no shortage of books on the historical Jesus in recent years, but if one restricts oneself to writers broadly sympathetic to the ‘third quest’ who have something interesting to say about Jesus’ miracles, four names in particular stand out: A.E. Harvey, E.P. Sanders, Gerd Theissen, and J.P. Meier.

According to Harvey, miracle working is attested of Jesus with a high degree of historical certainty. There are few stories from around the time of Jesus that exactly parallel the miracles of Jesus, and we should note the restraint of the Gospel stories. In the period from 200 bce to 200 ce the number of miracles recorded which are remotely comparable to those of Jesus is astonishingly small. In Judaism the only roughly contemporary and comparable figures are Ḥoni the Rain Marker and Ḥanina ben Dosa. There was, however, a Jewish concern to ban sorcery, which suggests that magic was perceived as a threat.

In Harvey’s view the Gospels lack the supernatural or psychic clairvoyance that seemed to be characteristic of Hellenistic miracle workers such as Apollonius. Moreover, Harvey argues, Jesus did not choose the style of a charismatic figure like Ḥoni in producing miracles by prayer. In carrying out exorcisms Jesus opted for a type of healing that could be dangerously ambiguous, but he refrained from employing means that looked magical.

Having explored the options allowed by the constraints under which Jesus operated, Harvey concludes that the miraculous activity of Jesus conforms to no known pattern in the ancient world. He draws attention to the mysterious nature of Jesus’ miracles and to the contest with the demonic powers. He points out that no fewer than eight of Jesus’ cures were of the deaf, the blind and the lame. According to Harvey such cures were unprecedented in Jewish culture (though not at pagan shrines such as those of Asklepios). The way to understand them, therefore, is to compare them with the prediction of the new age in Isa. 35:5–6. It is not so much that Jesus set out to fulfil this prophecy, as that both in the Isaiah passage and in Jesus’ day such conditions as blindness, deafness, lameness and demon-possession were seen as otherwise insuperable barriers to the attainment of the new age. Since it was Jesus’ task to act as God’s agent in inaugurating this new age, he performed such miracles as a sign of what he was about.

Harvey is one of the two scholars Sanders engages with in his discussion of miracles in Jesus and Judaism, the other being Morton Smith. Although Sanders regards Harvey’s work as a worthy attempt, he takes issue with his suggestion that Jesus ‘opted’ to perform mainly those miracles that would suggest the coming of the new age, partly through their being largely unprecedented in his culture, and partly through their correspondence to Isa. 35:5–6. Sanders notes that the miracles of Elijah and Elisha, though remote in time, undermine Harvey’s claim that Jesus’ miracles were unique in Jewish culture. But his principal objection is to Harvey’s supposition that Jesus deliberately selected what kind of healings to perform. To Sanders, ‘It seems much more likely that Jesus performed those miracles which came to hand.’ The ailments that Jesus is reported to have cured are just those that one might expect to be susceptible to faith-healing. In the end, then, ‘Harvey’s careful argument and its failure’ merely serve to convince Sanders that it is futile to seek to understand Jesus’ miracles through their partial correspondence to Isaiah 35.

Sanders finds himself more sympathetic to Morton Smith. He is persuaded by Smith’s argument that it was Jesus’ miracles, rather than Jesus’ teaching, that first attracted the crowds. He also finds merit in Smith’s case that Jesus was a magician, but ultimately regards this view as ‘unsatisfactory because it leaves largely out of account the persuasive evidence which makes us look to Jewish eschatology as defining the general contours of Jesus’ career’. The Gospel material suggests that ‘prophet’ would fit Jesus better than ‘magician’, though the miracles by themselves do not support the conclusion that Jesus was an eschatological prophet; they are merely consistent with it. Given the assumptions of the society in which Jesus operated, Sanders allows that it is also probable that both Jesus and his followers saw his miracles as signs of authentication by God, although the Gospels indicate that he did not perform them for this purpose. One is thus left with the impression, though Sanders does not quite say so in so many words, that Jesus’ miracles were mainly a crowd-pulling device.

Sanders is also concerned to correct what he sees as a number of common misconceptions, two of which bear directly on the question of Jesus’ miracles. The first of these is that exorcisms were generally regarded as signs of the kingdom, which Sanders regards as simply erroneous. The second is that Jesus himself regarded his healing and exorcism as signalling the in-breaking of the kingdom, which Sanders regards as possible but not probable. In particular he disputes the common interpretation of Mt. 12:28/Lk. 11:20 as a detached, authentic saying of Jesus that indicates that the kingdom has arrived in his exorcisms.

Curiously, Sanders appeals to just this verse towards the end of the more extended discussion of Jesus’ miracles in his later book, The Historical Figure of Jesus. Here Sanders argues that although ‘Jesus did not want to rest his case on his miracles’, his reply to John the Baptist ‘indicates that Jesus viewed them as proving that he was the true spokesman for God’. This reply recalls Isaiah 35 and suggests that Jesus claimed to be fulfilling the prophecy. But Sanders is now prepared to go even further than this:

More fully, he [Jesus] probably saw his miracles as indications that the new age was at hand. He shared the evangelists’ view that he fulfilled the hopes of the prophets—or at least that these hopes were about to be fulfilled.

Part of the evidence Sanders adduces for this is the eschatological sayings Jesus connects with his exorcisms, Lk. 10:17–18 and Mt. 12:28. Sanders has apparently changed his mind about the significance of such sayings; in particular, he now finds that the miracles did, unequivocally, have an eschatological significance for Jesus.

The discussion that leads up to this point in the later book suggests an increased emphasis on Jesus’ miracles compared with the previous one. In Jesus and Judaism Sanders included a 17-page chapter on ‘Miracles and Crowds’ comprising 5 per cent of the book; in The Historical Figure of Jesus the corresponding chapter on ‘Miracle’ contains 37 pages and represents nearly 13 percent of the whole. Much of the additional space is taken up with a general discussion of miracles and magic in the ancient world, in part aimed to correct the erroneous notion that Jesus’ miracles could be (now) or would have been (then) taken as evidence of his divinity. Sanders also includes a brief survey of other miracle-working figures from around the time such as Apollonius, Hanina and Ḥoni, the exorcist Eleazar, and the would-be prophet Theudas. He then turns to a brief treatment of the synoptic Evangelists’ portrayals of Jesus’ miracle working, and argues that Jesus was certainly an exorcist, even if the tradition has expanded on his exorcisms. Sanders again agrees with Morton Smith that Jesus’ miracles were the basis of his fame, but still finds it more helpful to regard Jesus as a prophet rather than a magician.

By this stage Sanders has already signalled his agreement with Cicero that anomalies are impossible. He therefore argues that the miracles of Jesus were probably rather minor and attracted little permanent interest, whereas the traditions behind the Gospels worked them up into something rather more spectacular. He would presumably agree with Theissen (see below) that the historical Jesus was indeed an exorcist and healer, while the nature miracles (or anomalies) were the product of post-Easter reflection, but he seems curiously reluctant to come out and say so plainly.

This distinction between the healings performed by the historical Jesus and the ‘nature miracles’ (not Theissen’s preferred term) attributed to him by post Easter faith is one of the key concerns in Theissen and Merz’s treatment of Jesus’ miracles in The Historical Jesus. Theissen and Merz are aware that at first sight the presence of seemingly impossible happenings (anomalies, in my terminology) in primitive strata of the Jesus tradition threatens to cast severe doubt on the historical reliability of the whole. They therefore argue in some detail that whereas Jesus’ exorcisms and healings are attested beyond reasonable doubt (being attested in multiple sources, including Josephus, and multiple genres), the same is not so of the ‘nature miracles’, which occur only as occasional, isolated stories, without being referred to in the summaries or sayings material. The original insight Theissen brings to all this is gained from developing the ideas presented in his earlier book on the miracle stories. He repeats part of his analysis of miracle story motifs and then goes on to classify the Gospel miracle stories into Exorcisms, Therapies, Norm miracles (e.g. healing on the sabbath), Gift miracles, Deliverance miracles and Epiphanies. He suggests that the first three belong to the ‘Effects of the historical Jesus’ whereas the second three display the ‘Presupposition of Easter faith’, and sets them out in a diagram showing how the six classes fall into three corresponding historical and post-Easter pairs.26

As in his earlier book, Theissen notes how many of the miracle stories reflect a popular trend that smoothes out what was characteristic of Jesus, but goes on to show how Easter experiences have been woven into the gift, rescue and epiphany miracles. He then concludes:

It is therefore no modern whim to note that the ‘nature miracles’ which we find hard to understand—the multiplication of loaves, the miracle with the fishes and the walking on the water, along with the transfiguration—have a special character within the miracle tradition. They are demonstrably steeped in Easter motifs, which cannot be said of the exorcisms and therapies.

In the course of his treatment Theissen gives examples of miracle stories from contemporary Jewish and pagan sources, and briefly discusses how Jesus compares with other contemporary miracle workers such as the θεῖος ἀνήρ, the magician, the rabbinic charismatic miracle-workers, and the Jewish sign-prophets. Theissen is unconvinced by Smith’s argument that Jesus was a magician; he accepts Vermes’s case for the existence of the charismatic holy men and of striking parallels between them and Jesus, but notes that there are also differences (Jesus, unlike these charismatics, did not effect his miracles through prayer but did give them eschatological significance); and he finds parallels between Jesus’ and the sign-prophets’ expectation of miracles in the future, but notes that the sign prophets were not said to perform healings and exorcisms in the present. Theissen accordingly concludes, ‘The uniqueness of the miracles of Jesus lies in the fact that healings and exorcisms which take place in the present are accorded an eschatological significance.’ This is fully in accord with the view already expressed in his earlier book on the miracle stories, which Theissen goes on to quote.

The final historical Jesus book to be considered here is the massive account of Jesus’ miracles given by John P. Meier. The bulk of this (393 pages) is taken up with a historical-critical analysis of each of the miracle pericopae, since Meier believes that it is useless to appeal to the miracle stories as evidence of Jesus’ miracle-working unless some of them can be shown to contain reliable historical tradition.30 Meier finds this to be the case for a minority of the exorcisms, healings and raisings of (supposedly?) dead people, but this is enough to establish his case. Most of the ‘nature miracles’ (a category Meier disapproves of) are found wanting, with the exception of the feeding stories, behind which Meier believes there is probably some historical kernel.

This analysis follows on from a discussion of whether Jesus performed any miracles at all. After reviewing the evidence Meier is firmly convinced that he did:

To sum up: the historical fact that Jesus performed extraordinary deeds deemed by himself and others to be miracles is supported most impressively by the criterion of multiple attestation of sources and forms and the criterion of coherence. The miracle traditions about Jesus’ public ministry are already so widely attested in various sources and literary forms by the end of the first Christian generation that total fabrication by the early church is, practically speaking, impossible. Other literary sources from the second and third generation—M, L, John, and Josephus—only confirm this impression. The criterion of coherence likewise supports historicity; the neat fit between the words and deeds of Jesus emanating from many different sources is striking …

The curious upshot of our investigation is that, viewed globally, the tradition of Jesus’ miracles is more firmly supported by the criteria of historicity than are a number of other well-known and often readily accepted traditions about his life and ministry.

This statement puts forcefully what virtually all the writers reviewed here accept. But not all scholars would agree. Harold Remus, for example, regards it as plausible but unprovable that Jesus was in fact a healer. Burton Mack goes further, taking the miracle stories to represent the myth of origins of one type of primitive Christianity, from which no conclusions may legitimately be drawn about any healing activity on the part of the historical Jesus.33 For the purposes of this study, however, Meier’s position will be taken as the ‘standard’ view, in the sense of a position with which many scholars agree and which may be used as a standard of comparison with the material to be examined.

Prior to this argument Meier devotes 82 pages (including endnotes) to the issue of ‘Miracles and Ancient Minds’, which includes a discussion of comparative material from Jewish and pagan sources. It is helpful that Meier sets out this material in some detail from the outset, and much of what he says is sound. One clear limitation of Meier’s discussion, however, is that it is quite deliberately one sided, since he is (quite properly) concerned to overturn prevailing views that tend to exaggerate the similarities between the Gospel accounts of Jesus and accounts of contemporary miracle-workers and magicians. Although this is a necessary exercise, it leaves the reader without any clear idea of what light may legitimately be shed on Jesus’ miracles from this comparative material.

Meier’s book contains one of the fullest treatments to date (in the context of a book on the historical Jesus) of the comparative Jewish material to be studied here. When one turns to works dealing more specifically with the context of Jesus’ miracles, however, one finds a curious reluctance to engage with this specifically Jewish material in any depth. R.M. Grant’s study, for example, contains much about changing beliefs in miracles in the Roman Empire as a whole, but gives remarkably scant attention to such major Jewish authors as Philo and Josephus. As the title of his book suggests, it is with Graeco-Roman rather than Jewish thought that Grant is primarily interested to compare Christian ideas of miracle. The same tends to be true of the two books on this subject by Howard Clark Kee.35 From these one may learn a great deal about the medical ideas of Galen or the healing cults of Asklepios and Isis, a fair amount about Pliny, Philostratus and Apuleius, a reasonable amount about views of miracle in the Old Testament, but virtually nothing about how miracle was perceived in late Second Temple Judaism, except as this was refracted by the New Testament authors. When Kee does come to discuss Josephus, his treatment of him is so one sided (looking solely at Josephus’s account of the portents of the fall of Jerusalem, where he resembles Roman historiography) as to be positively misleading. The merit of Kee’s work, on the other hand, is in his insistence that comparative material be taken in its proper context and not used anachronistically. This enables him to launch a stinging attack on religionsgeschichtlich approaches in general and Rudolf Bultmann in particular; third-century romances, for example, are not to be used to illuminate the historiography of first-century Evangelists.

One comparative category that has come under sustained attack in recent years is that of the θεῖος ἀνήρ, or ‘divine man’, to which the Evangelists or their traditions are supposedly meant to have conformed their image of Jesus in the miracle stories. It was partially against this tendency in New Testament scholarship that Vermes was reacting when he placed Jesus firmly within the context of the Judaism of his day. Attacks on the notion that ‘divine man’ was a pre-existing category to which Jesus could be conformed have been more notably launched, however, by scholars such as Otto Betz, David Tiede, Howard Clark Kee, Carl Holladay and Barry Blackburn.37 The works of Tiede, Holladay and to a lesser extent Blackburn contain substantial discussions of parts of Philo and Josephus, which often touch on these first-century writers’ treatment of miracle, and to that extent they are highly relevant to the present project; but, being restricted to the single issue of the ‘divine man’ they do not cover the entire field. Together, however, these authors build up an impressive case for dropping θεῖος ἀνήρ as a comparative category in discussing the miracles of Jesus.

A comparative work of a rather different nature employs semiotic approaches to attain religionsgeschichtlich ends; despite its forbidding title, Werner Kahl’s comparison of around 150 pagan, Jewish and Christian miracle stories makes several noteworthy points. Kahl’s starting point is that too many previous attempts to locate the miracle stories in their comparative religious settings have been unsystematic, undisciplined and far too subjective. To correct this Kahl proposes to subject a number of ancient stories of miraculous healing to a rigorous structural analysis, and then on that basis to judge whether those from the New Testament more closely resemble their Jewish or their pagan counterparts.

This is not the place to follow all the details of Kahl’s analysis, but we must note an important distinction he makes in the course of conducting it, not only in order to understand his conclusions, but also because it will prove useful to borrow both his distinction and his terminology. Kahl complains that the term ‘miracle-worker’ is too vague to be useful, since it is applied to a wide variety of figures whose function in miracle stories varies equally widely. He therefore proposes to divide ‘miracle-workers’ into Bearers, Petitioners and Mediators of Numinous Power:

Because of this diversity, I will refrain from using the term ‘miracle worker’ in my analysis, and introduce instead the terms ‘bearer of numinous power’ (BNP) for subjects who incorporate healing power in themselves, ‘petitioner of numinous power’ (PNP) for those whose function is to activate their gods through prayer, and ‘mediator of numinous power’ (MNP) for those subjects who mediate a BNP’s numinous power for the performance of a miracle.39

Thus, for example, throughout the Old Testament and much other Jewish literature, the BNP is always Yahweh. When the prayer of an Elijah or a Ḥoni causes drought or brings rain, the prophet or holy man acts as a PNP. When the Red Sea parts at the smiting of Moses’ rod, Moses acts as an MNP.

Armed with these distinctions, Kahl goes on to argue that the portrayal of Jesus as miracle-worker in the Gospels is unique for its time, though it bears some affinity with both Jewish and pagan parallels. Kahl finds the Q story of the healing of the centurion’s servant to be basically Jewish, the miracle stories in Mark to derive from a mixture of Jewish and pagan elements, and those in Matthew to be more nearly Jewish again. Here Kahl opens himself to the charge that he has not adequately accounted for the redactional activities of these two Evangelists, for if Matthew’s stories, mainly taken over from Mark, appear more Jewish, it may equally be that it was Mark or his more immediate tradition, rather than the originating point of his miracle stories, that gave them a more pagan feel in the first place. Kahl does, however, refer to Lukan redaction, for he finds that in Luke there is a tension between miracle stories that portray Jesus as the BNP and the Lukan redaction, which tends to turn him into an MNP like Elijah or Elisha. In Acts, however, Jesus becomes a transcendent BNP whose power is invoked by apostles such as Peter and Paul, who in turn are always portrayed as MNPs.

Kahl finds that the closest religionsgeschichtlich parallel to the Gospel presentation of Jesus is Philostratus’s Life of Apollonius of Tyana. But since this was written approximately 150 years after the Gospels, this leaves Jesus as virtually unique in being an immanent BNP (sometimes found in a pagan milieu) of whom a multitude of miracle stories is told (which is more typical of a Jewish milieu).

Although this survey has encompassed a number of disparate writers, several points of agreement emerge from what might be termed the ‘standard’ view, again in the sense of a view that may be used as standard for comparison. First, there is a consensus among virtually all the scholars reviewed here that Jesus did indeed perform healings and exorcisms that his contemporaries thought remarkable, and that this can be regarded as virtually certain. There is also a growing consensus that this miraculous activity formed an integral part of Jesus’ ministry, and should not be brushed aside to leave room for a Jesus who was almost entirely a teacher.

Thirdly, virtually all but the most conservative, scholars regard the so called ‘nature miracles’ (or what might be termed ‘hard anomalies’) to be the creations of the primitive church rather than deeds of the historical Jesus. The conservative case appears to rest on (valid enough) arguments for the logical possibility of anomalies together with tacit or explicit appeals to the authority of scripture or the probity of the Evangelists. The opposite case appears far stronger, however, since it can rely on a number of strands: (a) the problematic nature of alleged anomalies, which, though they cannot be strictly regarded as a priori impossible, are nevertheless highly improbable given our current state of knowledge about the world; (b) the post-Easter motifs that Theissen in particular notes as being characteristic of the ‘nature miracles’; and (c) the fruits of historical-critical analyses such as that of Meier, which show that few if any of the ‘nature miracle’ pericopae can safely be regarded as going back to the historical Jesus.

Fourthly, there is broad agreement that Jesus’ miracles, like the rest of Jesus’ ministry, are to be understood against the background of his eschatology. Fifthly, there is wide agreement that, pace Smith, Jesus is best understood not as a magician but as a prophet, and probably as an eschatological prophet. Sixthly, several writers call attention to the uniqueness of Jesus’ miracles, though this is understood in a variety of ways. For Harvey, the uniqueness lies principally in the absence of similar healing stories from the Judaism of the time, and in the fact that Jesus, unlike, say, Ḥoni, did not use prayer to work his miracles. For Kahl, the uniqueness is constituted by the fact that up to his time, Jesus alone is depicted as an immanent BNP to whom a whole series of healing miracles is attributed. For Theissen, Jesus’ uniqueness lay in using particular acts of healing in the here and now to have eschatological significance. For Meier, it lies not in the miracles per se but in their combination with other elements of Jesus’ ministry:

… the overall configuration, pattern, or Gestalt of Jesus as popular preacher and teller of parables, plus authoritative interpreter of the Law and teacher of morality, plus proclaimer and realizer of the eschatological kingdom of God, plus miracle-worker actualizing his own proclamation has no adequate parallel in either the pagan or the Jewish literature of the time.

These six areas of broad agreement may be accepted as working hypotheses for the purposes of this study, insofar as they supply a ‘standard’ view of Jesus to compare with his Jewish context. In what follows we shall focus on the Jewish context.


Studies of Jesus’ miracles, like those just reviewed, understandably focus on the New Testament evidence, thereby treating the Jewish context as mere background. In this study the key move will be to reverse the perspective, placing the Jewish context squarely in the foreground as the main focus of investigation while relegating the New Testament evidence for Jesus to the background. The purpose of this change of perspective is to enable Jesus’ miracles to be seen in their Jewish context, rather than viewing the Jewish context through the lens of Jesus’ miracles.

The danger with viewing the Jewish context through a Gospel lens is that one will almost inevitably end up with a distorted picture. If one sets out to write mainly about, say, the historical Jesus or his miracles then any discussion of Jewish context will be limited in scope and necessarily selective; the principle of selection will inevitably be one of perceived relevance to the main topic (i.e. Jesus), and so one will tend to end up with a collection of those snippets of Jewish background that most closely resemble or contrast with Jesus’ miracle-working. These may be quite unrepresentative of the views on miracle expressed in the surviving Jewish literature as a whole. A related danger (to which Meier is keenly sensitive) is that one may exaggerate the similarity between some aspect of Jesus’ ministry and some aspect of an apparent Jewish parallel and so end up distorting both. Vermes falls into both traps in his attempt to assimilate Jesus to a putative class of charismatic holy men exemplified by Ḥoni the Rain-Maker and Hanina ben Dosa. Several writers, Vermes included, fall into it when considering possible parallels to Jesus in the Qumran literature. Sanders and Theissen give useful accounts of Jesus’ miracles that are sensitive to the Jewish context, but cannot do full justice to the Jewish context within the format of their studies of the historical Jesus. Meier goes some way towards addressing this issue in an excursus, but even so it is only an excursus, and not a full study.

These criticisms cannot be levelled at studies that focus on particular aspects of Judaica relating to miracles, many of which will be engaged with in the course of this study. Here the problem is not one of distortion, but rather the lack of any recent study with the comprehensive objective of discussing miracle in Second Temple Judaism as a whole, as opposed to a limited range of texts. It is this lack that the present study aims to supply.

Clearly there is no such thing as the ‘Second Temple Jewish view of miracle’. The aim here must therefore be not to construct a false synthesis, but to map a range of diverse opinions. To carry out the task successfully, avoiding both a false synthesis and the temptation to view Jewish ideas on miracle through a Gospel lens, it is clearly better to approach the material text by text, or author by author, rather than thematically or according to any other scheme. The question is then which texts or authors to examine. To attempt to examine every single available Second Temple text that might have some bearing on the question of miracle would inflate this study beyond what is either reasonable or necessary and, in the case of the less fruitful texts, would result in a considerable amount of discussion that contributed very little. It is thus essential to be selective.

The two strongest candidates for inclusion are the works of Josephus and Philo, the obvious reason being that they are the two most substantial surviving corpora, the less obvious (until the matter is examined) being that they are the two authors most aware of miracle as an issue. Pseudo-Philo’s Liber Antiquitatum Biblicarum (henceforth LAB), the fragments of Artapanus, and the book of Jubilees provide further examples of re-written bible (found both in Josephus and Philo) that are also worth looking at from the point of view of miracle. The last of these, however, is closely related to the myth of the fallen Watchers in 1 Enoch, so this text should also be examined (if only for its demonology). The Qumran literature also contains one or two texts that appear to narrate miracles, and rather more texts that seem to be connected with demonology and exorcism, and these too should all be looked at. Although one might not ordinarily associate wisdom with miracle, the two main exemplars of Second Temple wisdom literature, Ben Sira and the Wisdom of Solomon, both have contributions to make on the question (largely through their rehearsal of biblical traditions), so these may also be included. The book of Tobit contains both a healing story and an exorcism, and this makes it another good candidate for discussion. Thereafter it is not clear what else may be profitably dealt with in detail. The Lives of the Prophets is greatly interested in miracles performed by the prophets, just as the Testament of Solomon is greatly interested in demons, but both texts are probably too late and too much under suspicion of Christian influence to be used as reliable sources for Second Temple Judaism. Apocalyptic writings such as 4 Ezra and 2 Baruch associate spectacular events with the end, but these are too remote from the Gospel miracles to be more than tangentially relevant to the present study. Several other texts contribute marginally to the issues of miracle and demonology, but to analyse them in any depth would be to risk encountering the law of diminishing returns. To keep this study within reasonable bounds, therefore, it will be best to confine it to the texts already listed, mentioning others only in passing.

Having decided the selection of texts to be studied, it is next necessary to decide the order and manner of treatment. The obvious place to start (given the relative abundance of their surviving writings) is with the works of Josephus and Philo, and these authors will accordingly form the subjects of Chapters 2 and 3. Both chapters will examine first the concept and language, second the distribution, and third the function of miracle in their respective authors. Following this firm start in Hellenistic Judaism (or, more specifically, in Greek-language Jewish texts consciously aware of the need to interact with their Hellenistic cultural environment) it makes sense to continue with a third text from that milieu, the Wisdom of Solomon, and to treat it in the same manner. This, together with the other wisdom text, Ben Sira, will thus be dealt with in Chapter 4. The movement from Wisdom to Ben Sira is also a movement from Greek-language Diaspora texts to Hebrew language Palestinian ones, a difference reflected in the loss of any self conscious reflection on miracle. It may be that Ben Sira represents the mainstream Palestinian Judaism (insofar as there ever was such a thing) of his time. Another author of which this might be said is Pseudo-Philo, whose LAB in some respects resembles the retelling of biblical history in Josephus and Philo. Pseudo-Philo’s LAB will accordingly be examined in Chapter 5, along the same lines as the previous literature.

The other main example of re-written bible originating in Hebrew is Jubilees, but at this point a change in the manner of treatment becomes necessary. On the one hand, Jubilees is clearly quite closely related to other Second Temple texts such as 1 Enoch and much of the Qumran literature. On the other, its treatment of miracle is too cursory to be worth examining in the same manner as before, whereas its views on demonology and eschatology demand rather more attention. In moving on to Jubilees one finds oneself moving into a stream of Second Temple Judaism distinct both from Hellenistic Jewish texts and from the putative Palestinian ‘mainstream’ perhaps represented by Ben Sira and Pseudo-Philo. For the sake of convenience this distinct stream may be called the ‘Enochic-Qumran’ stream (for reasons that should become clearer in the relevant chapters). As a first approximation, this is a stream of tradition closely influenced by the myth of the fallen Watchers in 1 Enoch (or by something very similar). To understand it, therefore, it is first necessary to examine this myth. Chapter 6 will accordingly deal with ‘Enochic Literature’, meaning mainly the Book of Watchers from 1 Enoch (together with the Qumran Book of Giants) and the book of Jubilees, although the conclusion will briefly attempt to trace some trajectories on into the Testaments of the Twelve Patriarchs. The following chapter, Chapter 7, will then examine a selection of Qumran texts (many of which seem related to Enochic literature) concerned with miracles or demons. Here again the nature of the material will dictate a substantial change in the manner of treatment. Virtually all the Qumran texts discussed are short and fragmentary, often with frustrating lacunae at critical points; these features all create particular problems of interpretation, which the reader may well experience as a sharp change of gear.

At this point only two texts out of those selected will remain: Tobit and Artapanus. These will be grouped together in Chapter 8 under the heading ‘Miracle and Romance’, although the two texts are in many respects quite different. Tobit is arguably related to the Enochic stream in some way, and was probably first written in Aramaic. With Artapanus, on the other hand, this study will have come full circle back to a Greek-language text from the Hellenistic Diaspora. It is, however, a rather different style of Hellenistic Jewish thought from that exhibited in the earlier texts.

The individual study of all these authors and texts may leave the reader with no clear overall picture of the Jewish context of Jesus’ miracles. It also leaves a substantial proportion of the Apocrypha and Pseudepigrapha untapped. In an attempt to correct both these shortcomings Chapter 9, ‘Miracle in Second Temple Literature’, will draw together the threads of the previous seven chapters by giving a thematic survey of what they have to say about the agents and functions of the miraculous, while at the same time introducing further material from some other texts that were not selected for more detailed examination.

Chapter 9 will thus conclude the literary survey of Second Temple Jewish views on miracle and their possible relation to the Gospel portraits of Jesus’ miraculous activity. The second (and, mercifully, shorter) part of the study will move on to a more historical approach, examining what this literature (together with some relevant rabbinic material) has to say about Jewish miracle-working contemporaries of Jesus. Since Vermes’s views on charismatic holy men have proved so influential, we shall begin by examining them in Chapter 10. Chapter 11 will go on to investigate the so-called sign prophets; and Chapter 12 will scrutinize the evidence for Jewish exorcists at around the time of Jesus.

In Chapter 13 the inquiry will shift from a historical to a social-scientific perspective. Although the social sciences cannot be used to fill in gaps in the historical data, they can at least prompt further questions (which may help show where gaps exist) and offer a fresh perspective. A full treatment of potential social-scientific insights into Second Temple Jewish miracles in general and Jesus’ miracles in particular would require a major study in itself; all that can be attempted in Chapter 13 is a sketch of how social anthropology may cast some light on Jesus’ activity as healer and exorcist in his social context. This will be conducted in three stages: first, a look at the healing role from the viewpoint of medical anthropology; second, a discussion of the relation between magic and miracle in Jesus’ Jewish context; and third, a brief appreciation of anthropological insights into spirit-possession and its bearing on Jesus’ ministry. Although brief, this chapter will act as a partial corrective to over-reliance on the literary products of urban elites. It will also suggest how Jesus could have been quite distinctive as a healer without being the only healer around, as well as indicating a rather different perspective on spirit-possession and exorcism in Jewish society from that suggested by purely literary and historical approaches.

This increased focus on Jesus will continue into Chapter 14, the Conclusion. The final chapter will not attempt to provide a detailed reassessment of Jesus’ miracles, but will rather indicate how the survey conducted in the previous 12 chapters might bear on the question. Many of the working hypotheses arising from the literature survey will be found to have been confirmed, though they will often have been qualified in the process. In general terms it will have been shown that Jesus—both the narrative Jesus of the Gospels and the historical Jesus—stands out as a highly distinctive figure in respect of his miracle-working in the Judaism of his time, and yet makes sense as a miracle-worker in his Jewish context. Stated thus baldly, this conclusion will inevitably appear either paradoxical or jejune. The interest lies in the detail.

Chapter 2

Miracle in Josephus


The Jewish historian Josephus provides the most obvious starting-point for an investigation into first century Jewish views on miracle. His extensive retelling of biblical narratives shows how the biblical miracles were understood by some Jews of his time, and his account of post-biblical history gives us the fullest indication we have outside the New Testament of first-century Jewish willingness to believe in contemporary miracles and miracle-workers. Although Josephus’s social class, background and experiences give him an outlook very different from that of the Gospels, he shares with them strong Jewish roots overlaid with Hellenistic influences. This chapter aims to give a general overview of Josephus’s approach to miracles; specific passages of interest will be treated in more detail as they arise in Chapters 10, 11 and 12. How far Josephus may be regarded as typical of the Judaism of his day will emerge in the chapters that follow.

The Concept and Language of the Miraculous in Josephus

No more for Josephus than for the biblical authors does miracle automatically entail our concept of anomaly. Josephus occasionally stresses that a miracle is outside the ordinary course of nature (e.g. Ant. 4.48–53). Moreover, he occasionally suggests that a natural explanation can rob a sign of its miraculous significance (Ant. 8.243–45). But many of the events Josephus regards as miraculous (for example the freak storm that attests Samuel’s warnings at Ant. 6.92) are not obviously exceptions to the laws of nature, whereas, as will now be shown, a more ‘biblical’ definition of miracle fits Josephus nicely.

This definition is in three parts. Miracles are, first, strikingly unexpected events that, secondly, are taken by believers to be, thirdly, acts of God. It will now be demonstrated that each of these parts applies to the miraculous in Josephus.

First, Josephus frequently calls attention to the strikingly unexpected nature of miraculous events. For example, at Ant. 4.48 Moses asks God to remove the rebels by no common means, and what then happens to them at 4.51–53 is clearly extraordinary, just as the fire which subsequently engulfs Korah and his companions is described as something unprecedented in human history (4.55). Similarly, Samuel’s speech to the Israelites who were demanding a king makes it clear that God is about to signal his displeasure in an exceptional manner: ‘For that which not one of you ever saw befall here before—a tempest at midsummer—that through prayer to God I shall cause you now to witness’ (Ant. 6.91). Similarly, the plague of hail is described as not only unprecedented by Egyptian standards, but larger than any hail known in cold climes (Ant. 2.305). At Ant. 2.303 the beasts sent to infest the country are said to be ‘the like of which no man had ever encountered before’. God’s help in Joshua’s battle against the Canaanites included sending hail ‘of more than ordinary magnitude’ and prolonging the day beyond its ‘customary measure’ (Ant. 5.60–61). When Elijah restores the widow’s son to life this is described as ‘beyond all expectation’ (παρὰ πᾶσαν προσδοκίαν, Ant. 8.327).

Two passages are particularly pertinent. The first occurs in the preface to the Antiquities, where, setting out the main moral lesson to be learned from his history, Josephus remarks that those who conform to God’s will and obey his laws ‘prosper in all things beyond belief (πέρα πίστεως)’ while a contrary fate awaits the disobedient (Ant. 1.14); God’s providence thus manifests itself in extraordinary events and reversals. The second concerns the sign Hezekiah requests to confirm that he will recover from his illness. Hezekiah asks that the sun’s shadow should retrace ten steps because ‘things that are beyond belief (τὰ … παράλογα) and surpass our hopes are made credible by acts of a like nature’ (Ant. 10.28). If a sign is to attest something hard to believe, it must itself be something extraordinary.

But there is one important qualification. Although the strikingly unexpected often forms part of Josephus’s concept of miracle, not everything he attributes to divine intervention or providence is strikingly odd. Towards the end of the Jewish War, for example, Josephus describes how a change of wind favoured the Roman attack on Masada, and suggests that God was thereby aiding the Romans (War 7.317–19, 332). Even less spectacular is Josephus’s ascription to divine providence of his sighting of another ship after his own foundered on the way to Rome (Life 15). Josephus’s concept of πρόνοια threatens to include all that happens; conversely, he draws no sharp line between the fully miraculous and the merely providential.

The second part of the proposed definition concerns the believer’s reception of miracles. Here the term ‘believer’ may refer either to characters within Josephus’s narratives or to Josephus’s readers.

The recurrent formula to the effect that readers are welcome to their own opinion on purported miracles (e.g. Ant. 1.108; 2.348; 3.81; 4.158; 10.281; 17.354) is not an indication of Josephus’s own doubt, but rather a stereotyped nod in the direction of his (possibly sceptical) Hellenistic readers. There are several places, such as his polemical passage against Epicureanism (Ant. 10.277–81) where Josephus makes it perfectly clear where he stands. Nevertheless, the formula recognizes the possibility of doubt, and therefore the need for belief on the part of the reader to accept the miracles as genuine.

The belief of characters in the narrative is also sometimes at issue. At Ant. 8.243–45 Josephus describes how a false prophet persuaded King Jeroboam that the visitor from Judah was not, after all, a genuine prophet. This is an addition to the biblical narrative, presumably intended to explain why Jeroboam persisted in his evil ways despite being given a sign. The Judaean prophet (to whom Josephus gives the name Jadon) comes to Bethel just as Jeroboam is about to offer a sacrifice. He prophesies Josiah’s future desecration of the altar, and goes on to offer this sign in support: ‘The altar shall be broken in an instant and all the fat of the victims on it shall be spilled upon the ground’ (Ant. 8.232). Not only does this duly take place, but Jeroboam’s hand becomes paralysed when he stretches it out to order Jadon’s arrest. Subsequently, however, a false prophet argues that these supposed signs have a natural explanation: Jeroboam’s hand had been temporally numbed by the fatigue of carrying the sacrifice, and the new altar had collapsed under the weight of the things placed upon it. That Josephus has Jeroboam accept these rationalizations suggests that he recognizes a certain ambiguity in miraculous signs: to see them as acts of God one must be willing to believe.

This can be so even when the events witnessed are quite exceptional. When Moses shows his miraculous signs to Pharaoh (Ant. 2.284–87) all the Egyptian king sees is a competition between rival magicians; so far as Pharaoh is concerned, Moses was simply ‘trying to impose on him by juggleries and magic (τερατουργίαις καὶ μαγείαις)’. The one thing Pharaoh refuses to see in the signs Moses performs is an act of God.

This passage also illustrates the aptness of the third part of the definition: for Josephus, a miracle is essentially an act of God. This is precisely the point at issue between Moses and Pharaoh, as Josephus makes abundantly plain when he has Moses say

Indeed, O king, I too disdain not the cunning of the Egyptians, but I assert that the deeds wrought by me so far surpass their magic and their art as things divine are remote from what is human. And I will show that it is from no witchcraft or deception of true judgement (γοητείαν καὶ πλάνην τῆς ἀλγθοῦς δόξης), but from God’s providence and power (θεοῦ πρόνοιαν καὶ δύναμιν) that my miracles proceed. (Ant. 2.286)

This point can be illustrated from many other passages. At Ant. 2.274 the purpose of the signs Moses is given to perform is to convince everyone that he has been sent by God. When Moses strikes the sea and it recoils, this is said to be a clear manifestation (ἐπιφάνειαν) of God (Ant. 2.339). When he subsequently makes a pool of brackish water drinkable, he first prays to God and then impresses upon the Hebrews that it is God who has ‘promised to render the water such as they desired’ (Ant. 3.7). Likewise the water Moses produces by striking a rock is described as a boon from God (3.35). The earthquake that engulfs Dathan’s company ‘furnish[es] an exhibition of God’s mighty power’ (4.52), just as the flame that consumes Korah is one such ‘as might be kindled at the bidding of God’ (4.55). Indeed, the speech Moses makes prior to this incident dwells at some length on God’s previous interventions culminating in the prayer, ‘Prove now once again that all is directed by thy providence, that nothing befalls fortuitously (αὑτομάτως), but that it is thy will that overrules and brings everything to its end’ (4.47).

The same insistence on the divine origin of miracles recurs in connexion with other figures. It is God who co-operates with Joshua by means of thunderbolts in his battle against the Canaanites (Ant. 5.60); it is God who heeds Samuel’s prayer and vexes the Philistines with earthquake, thunderclaps, and fiery lightning (6.27), just as it is God’s displeasure that is manifested by the freak midsummer storm (6.92); it is God who manifests his favour by sending fire from heaven to consume Solomon’s sacrifice (8.118–19) as he later consumes Elijah’s on Mount Carmel (8.337–43). Similarly, it is God who fills the vessels that Elisha instructs the widow to borrow (9.48–50), God who visits a pestilential sickness upon Sennacherib’s army (10.21), and God who shows Hezekiah the sign of the reversing shadow at Isaiah’s request (10.29). Further examples occur in connexion with Daniel (10.258–62), Daniel’s three friends (10.214–15), Onias (14.22) and even the Roman legate Petronius (18.284–85). In Kahl’s terminology, for Josephus the BNP is always God.6

All this indicates that events in Josephus that we should call ‘miracles’ conform to the proposed tripartite definition: strikingly unexpected events that believers perceive as acts of God. But since Josephus uses no single Greek term to refer to all these events, it would be premature to call this his concept of ‘miracle’. One must first investigate the vocabulary he employs to refer to such events.

The two words that Josephus uses most frequently in connexion with miracle are σημεῖον and παράδοξος. In the Antiquities παράδοξος is used up to 20 times in connexion with the miraculous (Ant. 2.223, 295, 345; 3.1, 30, 38; 5.28; 10.214 and perhaps also 2.216, 267, 285; 3.14; 9.14, 58, 60, 182; 10.24, 28; 13.282; 18.63). Although it is not used in a miraculous sense in Josephus’s other writings, both there and in its non-miraculous uses in the Antiquities it nearly always means ‘amazing’, ‘astonishing’ or ‘extraordinary’ rather than merely ‘surprising’. On the other hand, θαῦμα, θαυμάσιος and θαυμαστός, which are also applied to what is wonderful, marvellous or extraordinary, are virtually never used by Josephus in connexion with miracles (the exceptions being Ant. 2.232, 265 and 9.182; in the last two of these the θαυμα- word is used in conjunction with τέρας and παράδοξα respectively). It might be too much to say that Josephus observes a sharp distinction between θαῦμα and παράδοξον (or occasionally παράλογος), reserving the latter to mean ‘miracle’, but there is certainly a tendency in that direction.

In a number of instances Josephus appears to use παράδοξον virtually as a substantive meaning ‘miracle’. A particularly striking example is the opening of Antiquities Book 3 where it appears as the first word, and is thus made to bear the full theological weight of the account of the Red Sea crossing with which Book 2 concluded.

To be sure, of the 20 instances of a miraculous use of παράδοξος in the Antiquities, over half are only possibly miraculous. Yet for each of the 12 dubious instances a case can be made for a miraculous connotation. For example at Ant. 2.267 the burning bush is described as an ὄψιν παράδοξον, in a context which would certainly allow, even if it does not demand, the translation ‘miraculous spectacle’. At 9.14 we are told that Josaphat ‘rejoiced at the wonderful way (ἐπὶ τῷ παραδόξῳ) in which God had helped’ him win a victory (by sending fear and confusion upon the Ammanites so that they slaughtered one another). If this is not quite a miracle, it is something very close to one. Even at 10.28 where, in the first instance, it is Isaiah’s news that Hezekiah will recover from his illness that is described as παράδοξον, one could nevertheless argue that since the king appears to believe that it will take a miracle to cure him, it is effectively a miracle that is in view.

If such arguments are accepted in the majority of the dubious cases, this will show that παράδοξος was one of Josephus’s key miracle-words, and this in turn reinforces the argument that the extremely surprising was part of Josephus’s concept of miracle.

Although σημεῖον is used in many senses that have nothing to do with miracle (in Ant. 19.29–256, for example, it is used 13 times to mean ‘password’, and elsewhere it may mean ‘signal’, ‘mark’, ‘trace’, ‘signification’, ‘symbol’, ‘legionary standard’ etc.), it is used in connexion with the miraculous up to 18 times in the Antiquities (2.274, 276, 280, 283, 284, 327; 6.91; 8.232, 236, 244; 10.28, 29 bis; 20.168; and perhaps also 6.54, 57, 93; 19.9) and up to 10 times in the Jewish War (1.23, 28, 332; 2.259; 4.623; 6.285, 296, 315; 7.438; and possibly 3.404). With the exception of the block in Antiquities 19 where it means ‘password’, this arguably makes the miraculous use the single most common use of σημεῖον in the writings of Josephus, as well as making it one of Josephus’s most characteristic words for miracle.

The question then arises whether σημεῖον has any special nuance for Josephus (when used of the miraculous). The answer is that it does, but the nuance varies between the War and the Antiquities.

In the War σημεῖον (in its miraculous connotation) tends to mean ‘omen’ or ‘portent’. At War 1.28 and 6.296, 315, for example, it is used of the portents that preceded the fall of Jerusalem. At 1.23; 3.404 and 4.623 it refers to the omens of Vespasian’s elevation to the purple. When Herod narrowly escapes from a collapsing building he sees this as ‘an omen (σημεῖον) alike of perils and of preservation during the coming campaign’ (War 1.332). The only exceptions to this usage occur at War 2.259; 6.285 and 7.438, all of which refer to signs offered by false prophets (in the first case ‘signs of freedom’, in the second ‘signs of salvation’, and in the third ‘signs and apparitions’). These signs could be portents, but the context suggests that a divine saving action is in view.

It is, however, in the Antiquities that Josephus develops his most distinctive use of σημεῖον. The Septuagint uses the phrase σημεῖα καὶ τέρατα to refer to the whole complex of miraculous events associated with the exodus. But apart from one exception (which will be discussed below) in his exodus account Josephus confines the term σημεῖον to the three signs God gave Moses on Mount Sinai (the staff that transforms into a serpent, the hand that turns white, and the water that becomes blood). God instructs Moses to use these signs to convince all men that God is with him and that he is acting in obedience to God’s command (Ant. 2.274). We are subsequently told that Moses finds these signs serviceable for this purpose (2.276) though he has to perform them and not merely describe them (2.280). Moses goes to Pharaoh and tells him about the signs he has been given (2.283), and, when Pharaoh mocks, proceeds to perform them again (2.284). On the other hand, when Josephus narrates the plagues, he does not call them ‘signs’. The word occurs only once more in the context of the exodus, and that is at 2.327, the exception that must now be considered.

At this point in the narrative, the Hebrews are camped next to the Red Sea, trapped and helpless before the pursuing Egyptians, and we are told that ‘they turned to accusing Moses, forgetful of all those miracles (σημείων) wrought by God in token of their liberation’ (so Thackeray translates, though πρὸς τὴν ἐλευθερίαν might more naturally be translated ‘for their liberation’ rather than ‘in token of’ it). Here the phrase is most naturally read as a reference to the plagues. It may be that, in retrospect, these plagues have also become ‘signs’ in that they authenticate Moses’ mission,12 but this explanation seems forced. It is preferable to admit that Josephus’s usage is not entirely consistent. But apart from this exception there is a distinct tendency in Josephus’s account of the exodus to restrict the term σημεῖον to miracles that authenticate God’s prophet.

This tendency continues as far as Ant. 10.29. At 8.232, 236, 244 σημεῖον is used of the signs given by Jadon at Bethel, and here the issue is precisely whether the apparent signs are genuine miracles from God authenticating the prophet’s message. At 10.28, 29 Hezekiah asks Isaiah for a sign to verify that his promise of recovery really is from God. At 6.54, 57 Samuel offers the newly anointed Saul three signs that God has indeed elected him to be king of Israel. Although these signs are perhaps not quite miraculous, they nevertheless serve to authenticate the prophet’s message. At 6.91 Samuel prevails upon God to show his displeasure through σημείων, to confirm what he has just been saying.

One doubtful case occurs at 8.347, where Jezebel is said to learn of the σημεῖα wrought by Elijah; this presumably refers to the fire which fell from heaven to consume his sacrifice on Mount Carmel. It could, perhaps, be argued that this was a sign authenticating Elijah as a true prophet, but this would be stretching a point, since neither Josephus nor the Bible represents this as the main point of the miracle. Samuel’s exhortations at 6.93 provide an even clearer exception, since in this context the signs of God to which he refers are far more likely to be his saving acts than miracles authenticating prophets. There is another possible exception at 1.332 where Jacob regards his defeat of the angel as a ‘σημεῖον of great blessings to come’, although one could reasonably argue that ‘sign’ does not bear the meaning ‘miracle’ here and so does not constitute a genuine exception.

These few exceptions apart, throughout the first nine and a half books of the Antiquities Josephus restricts the miraculous use of σημεῖον to the special sense of ‘authenticating sign’. From 10.234 onwards, however, this special usage drops out of sight, and Josephus reverts to that of the War. At 10.234 (in the story of the writing on the wall) σημεῖα appears to mean ‘omens’ or ‘portents’, as it does at 18.211 and 19.9, 94. At 20.168 certain ‘impostors and deceivers’ promise ἐναργῆ τέρατα καὶ σημεῖα to those who follow them into the desert.

The other significant word in Josephus’s miracle terminology, though it is not nearly so frequent, is ἐπιφάνεια. Its use in connexion with miracles is restricted to the Antiquities, but there it generally means a manifestation of God’s presence or of his saving power. The most notable occurrence of the latter is at Ant. 2.339, where the recoiling of the Red Sea is described as a ‘clear manifestation of God’. It has a similar sense at 9.60, where following Elisha’s miraculous capture of the Syrian army, ‘Adados was amazed at the marvel and at the manifestation of the God of the Israelites and His power.’ These are the only two instances which fit the idea of saving power (unless one also counts the reference to the ἐμφανείας τοῦ θεοῦ at 15.425). Neither the ‘divine manifestation’ of fire falling from heaven to consume Solomon’s sacrifice (8.119) nor God’s ἐπιφάνεια in the unexpected rainfall that greets Petronius’s decision to defy Caligula (18.286) are exactly saving acts, but rather signs of God’s approval. Although Thackeray’s translation does not bring out the fact, Josephus may intend to imply that Isaac’s fortuitous meeting with Rebecca (1.255) was due to a divine ἐπιφάνεια, though this would a low-grade miracle. It is a moot point whether ἐπιφάνεια at 3.310 should be translated ‘manifestation’ (meaning that the cloud above the tabernacle is a miracle produced by God) or ‘presence’ (meaning that the cloud symbolizes the presence of God).

Yet if Josephus’s use of ἐπιφάνεια fails to conform to any neat pattern, its use in connexion with miracle does serve to indicate once again that the notion of God’s action in the miraculous is important to him.

These three are the most significant words in Josephus’s vocabulary of miracle. He occasionally uses τέρας, but he is sparing of its use in the Antiquities (it occurs only at Ant. 2.265; 4.43, 291; and 20.168, although he also uses τεράστιον at 10.28, 232). He uses τέρας more frequently in the War, occasionally in the sense of ‘miracle’ (War 1.331; 5.411) but more often in the sense of ‘portent’ (1.28, 377, 378; 4.287; 6.288, 295). Although Josephus sometimes speaks of miracles as manifestations of God’s δύναμις he never uses δύναμις in the sense of ‘mighty work’ as a term for miracle.

This examination of Josephus’s miracle vocabulary can only take one so far. To understand what significance these words had for Josephus one must investigate what function miracles performed in his narratives. But before that task is attempted, it will be convenient to review the distribution of the miraculous material in Josephus’s writings.

The Distribution of Miracles in Josephus

Most of the examples in the previous section were taken from the Jewish Antiquities, and most of those came from the first ten books. This reflects the distribution of miraculous material within Josephus’s works. This section will explore this distribution further, first of all in relation to the Antiquities, and then with respect to the post biblical miracles in Josephus and the miraculous in Josephus’s other works.

As noted in the previous section, it is not always easy to distinguish the truly miraculous from the merely providential in Josephus, since he tends not to draw a sharp line between the two. This makes the distribution of miracles hard to quantify, since there is often a subjective element in deciding whether or not any given incident is to count as miraculous. For example, accounts of God’s punitive actions are scattered throughout the Antiquities. When the punishment is on the scale of Noah’s Flood (Ant. 1.75–103) or the plagues of Egypt (2.294–313) there can be little hesitation in seeing a miracle. In other instances one may be less sure. The king of Egypt took a fancy to Abraham’s wife, but ‘God thwarted his criminal passion by an outbreak of disease and political disturbance’ (1.164). In the wilderness period, God launched a pestilence upon those who consorted with the Midianite women (4.155). After being lured into disobedience by the false prophet, Jadon was killed by a lion ‘in accordance with the will of God’ (8.241). A sudden stroke from God seized the high priest Alcimus as he was about to pull down the wall of the holy place (12.413). God sent a mighty and violent wind to destroy the crops as a punishment for the death of Onias and for cheating the priests out of their sacrifices (14.28). King Herod’s illness worsened as God punished him for his misdeeds (17.168). Are such incidents miracles or merely examples of God’s retributive providence?

On the one hand, one might discount them as miracles on the grounds that there is nothing particularly extraordinary about them; they are all perfectly natural occurrences, and thus fail to meet the criterion of being strikingly unexpected events. On the other, one might argue that the context makes them miraculous, since it is the wicked who are punished, and a world in which misfortunes regularly fall upon those who deserve them is sufficiently contrary to our experience to count as strikingly unexpected. But perhaps that is the point: the story world of the Antiquities is one, Deuteronomically inspired, in which people tend to get their just deserts, so that the punishments just listed are not so much miracles as part of the regular workings of the system. Yet Josephus relates some spectacular punishments (such as that of Dathan, Ant. 4.51) which are unmistakably miraculous, so that it then seems arbitrary to exclude less spectacular punishments.

The difficulty of quantifying the miraculous in Josephus thus remains. Perhaps the best one can do is to work with both a wider and a narrower conception of miracle and discuss the range of figures that results from counting both ways. On that basis the proportion of miracles occurring in the first ten books of the Antiquities is between 84 and 94 per cent of the total for the whole work. There is a tendency for the miracles to cluster round the great prophetic figures: miracles associated with Moses, Samuel, Elijah, Elisha and Daniel together account for between 60 and 70 per cent of the total, with Moses taking a substantial proportion (around 36 to 43 per cent of the whole).

This largely reflects the distribution of miracles in Josephus’s main source, the Jewish Scriptures, since, on the whole, he reproduces the substance of his source with few omissions or additions. There are a few miracle stories he omits from his account of Moses, such as Miriam’s temporary leprosy (Num. 12:9–16) and the bronze serpent that saved the people from snake bite (Num. 21:4–9). There are also one or two he omits elsewhere, such as the consumption of Gideon’s unleavened cakes and the sign of his wet and dry fleece (Judg. 6:19–24, 36–40; cf Ant. 5.213–215). But the most extensive omission of miraculous material comes in his treatment of Elisha.

Josephus first introduces Elisha at Ant. 8.352–54, when he becomes Elijah’s disciple, but Elisha’s deeds are narrated at 9.33–183, at the end of which we are told that ‘he was a man renowned for righteousness and one manifestly held in honour by God; for through his prophetic power he performed astounding and marvellous deeds (θαυμαστὰ γὰρ καὶ παράδοξα ἔργα), which were held as a glorious memory by the Hebrews’ (9.182). Yet among Elisha’s ‘astounding and marvellous deeds’ Josephus makes no mention of his healing the well at Jericho (2 Kgs 2:19–22), cursing the jeering boys who are then torn by bears (2 Kgs 2:23–25), resuscitating the Shunammite woman’s son (2 Kgs 4:11–37), rendering edible the contents of a poisoned pot (2 Kgs 4:38–41), feeding a hundred men with five barley loaves (2 Kgs 4:42–44), healing Naaman of leprosy (2 Kgs 5), or recovering a sunken axe head (2 Kgs 6:1–7). It cannot be that Josephus accidentally turned over several pages of his source at once (as one might almost suspect had he been a modern author), since he is most unlikely to have been reading the Jewish scriptures in codex form. It is thus something of a puzzle why he should have omitted quite so many miracles from the story of someone he explicitly states to have been so renowned for his miraculous deeds, and whose exploits he introduces by referring to ‘the prophet Elisha, whose acts I wish to relate—for they are glorious and worthy of record—as we discover them in the sacred books’ (Ant. 9.46).

One can understand why Josephus chose to omit the story of the bears, since it hardly creates an edifying picture of Elisha’s temperament, and one might conjecture that he regarded some of the other stories (for example the axe head) as too trivial, or else as too repetitious of what he had already narrated in connexion with Elijah. But this still does not explain the exceptional extent of the omissions.

One must first attend to the stories that Josephus retained from the Elisha cycle in order to ascertain his principle of selection. Josephus opens his account of Elisha’s deeds with a lengthy narration of how the prophet interceded with God to provide the combined armies of Israel and Judah with urgently needed water and to assure them of victory over Moab, ‘for the sake of J[eh]osaphat who was a holy and righteous man’ (Ant. 9.33–44). He next narrates how Elisha came to the aid of Obadiah’s widow by providing her with a miraculous supply of oil, thereby freeing her from her creditors (9.47–50). Here Josephus reminds us that ‘her husband had saved the lives of the prophets who were to have been slain by Achab’s wife Jezabela’ (9.47). here follows a lengthy account of how Elisha assisted King Jehoram of Israel with information about Syrian plans, how Benhadad of Syria attempted to capture him, and how instead Elisha miraculously captured the Syrian army sent to arrest him (9.51–60). The section that follows (9.61–86) deals with the Syrians’ next campaign and their siege of Samaria, during which Jehoram failed to take Elisha’s life, and Elisha made a number of improbable predictions that nevertheless came true. Then, when the Syrian king fell ill and sent his trusted servant Hazael to enquire of Elisha, Elisha tearfully predicted Benhadad’s death and Hazael’s succession—tearfully, because he knew what damage Israel would suffer at Hazael’s hands (9.87–94). Elisha next appears when he sent one of his disciples to anoint Jehu as king, whereupon Jehu launched a successful coup d’etat and eliminated the house of Ahab along with the worshippers of Baal (9.106–39). After an interlude to catch up with affairs in Judah, Elisha makes one final appearance in which he predicts that Jehoram’s son Joash will win three victories against the Syrians, in accordance with the number of arrows he shot from his bow (9.177–81). Almost immediately thereafter Elisha dies, but contact with his body revives the corpse of a man murdered by robbers, thereby demonstrating that even the dead Elisha retains his numinous power (9.182–83).

What stands out from this summary is that in nearly all this Elisha material Josephus is concerned with Israel’s military and political affairs, in particular its continuing war against Syria and its change of dynasty. This suggests that Josephus selected the material most likely to be of interest to Graeco Roman readers, who would have expected a historical work to include such things rather than inconsequential stories of floating axe heads and poisoned pots. Yet this cannot be the whole story, since it does not explain the inclusion of the widow’s miraculous cruse of oil. Here Josephus stresses that her late husband was instrumental in rescuing the prophets of Yahweh from the machinations of Jezebel. The aid Elisha brings to her impecunious plight thereby exemplifies God’s retributive providence, reversing her misfortune as a reward for her husband’s faithfulness. This is a feature that most of the omitted miracle stories lack. The well at Jericho, the raising of the Shunammite woman’s son, the poisoned pot, the feeding of one hundred men, and the floating axe head are all good deeds performed by Elisha but they serve no more to illustrate God’s πρόνοια punishing the wicked and rewarding the righteous than they do to advance the military or political action. On the other hand, the stories that Josephus has included have the theme of God’s retributive providence at least running in the background: Elisha promises God’s help against Moab for the sake of the righteous Jehosaphat, the house of Ahab and the worshippers of Baal receive their due punishment, and Joash is semi rewarded for being only semi obedient (9.181).

One may wonder whether Josephus employed the same principle of selection in regard to the Elijah stories, from which he has omitted far less. Here, however, most of the biblical stories would have met one or other of his selection criteria, for either they further the account of Elijah’s struggle against the Baal cult (e.g. 8.337–42) or else they demonstrate God’s providential care for his prophet and his helpers (8.319, 321–23, 325–27, 349; 9.23). The revival of the widow’s son (8.325–27) and the fire from heaven (9.23) further serve to authenticate Elijah as a true prophet from God.

How Josephus’s miracle stories serve his narrative purpose will be discussed further in the next section. Enough has been said both to indicate that in rewriting biblical history Josephus broadly reproduces the distribution of miracles he finds in his source, and to offer a plausible explanation for his main departure from this practice. The next task is to explore the incidence of post biblical miracles in Josephus. Here the Antiquities continues to be the most important source, although the Jewish War also contains important material.

Post-biblical miracles in Josephus are sparse, but they are less sparse in the Antiquities than elsewhere. There are few impressive post biblical miracles even in the Antiquities. At 15.425 it is said that during the 18 months it took to build Herod’s temple it rained only at night. This must count as strikingly unexpected, as Josephus acknowledges with his apologetic statement that ‘this story, which our fathers have handed down to us, is not at all incredible if, that is, one considers the other manifestations of power given by God’. At 16.182 two of Herod’s bodyguards are destroyed by a flame that engulfs them on entering David’s tomb. At 18.285 God sends an unexpected shower of rain in support of Petronius and the Jews protesting against Caligula’s order to have his statue erected in the Jerusalem temple; this should be regarded as a miracle since Josephus stresses that the shower came out of a clear sky during a prolonged drought (compare the angry storm that backed Samuel’s protest against the institution of a king, 6.92). We are also told that Onias once prayed during a drought and that God sent rain as a result (14.22), so this may fall into the same category. A more controversial instance is the notice that Jesus Christ was a doer of παραδόξων ἔργων (18.63), which in view of Josephus’s usage elsewhere is almost certain to mean ‘miracles’.

None of these instances appear in the parallel sections of the War. What the War does share with the Antiquities is a number of less than miraculous providential interventions (for example the rain that falls just in time to prevent Herod’s brother having to flee from Masada [War 1.286–87 and Ant. 14.390], or Herod’s providential escape from the building that collapses at Jericho [War 1.331 and Ant. 14.455]), though even here the Antiquities reports more such incidents. The War and the Antiquities also share a number of prophecies and dreams, which are one of the chief manifestations of numinous power in Josephus’s account of post biblical history (e.g. Judas the Essene’s prediction of Antigonus’s death at Strata’s Tower [War 1.78–80 and Ant. 13.311–13], and Glaphyra’s dream of her husband just before her death [War 2.116 and Ant. 17.349–53]).

They also have in common a number of reports about ‘sign prophets’, but since these will be dealt with in Chapter 11 they need not be examined here. In any case, these sign prophets presumably failed to deliver on the promises Josephus attributes to them. On the evidence Josephus gives, they cannot be included among the workers of post biblical miracles.

The account of the false prophet at War 6.285–87 is immediately followed by a flash back to a sequence of seven portents which, in Josephus’ s view, should have warned the people of their impending fate, if only they had interpreted them rightly. A star resembling a sword stood over the city, a comet lasted a year, a bright light shone round the altar at night, a sacrificial cow gave birth to a lamb in the temple court, the massive brass gate of the inner court opened by itself, armies were seen riding through the skies, a voice from the temple announced, ‘We are departing hence’, and a prophet named Jesus ben Ananias walked the streets of Jerusalem with a message of doom from four years before the war (6.288–309). It is to the last of these, Jesus ben Ananias, that Josephus gives the most attention (6.300–309), perhaps because he sees this Jesus as an outstanding example of a true prophet in contrast to all the false ones. In any case, this prophet supplies the only part of the sequence that is typically Jewish; the portents that precede are more typical of Roman historiography (Tacitus includes four of the same portents in his account of the fall of Jerusalem at Histories 5.13). It maybe that Josephus included them with his Roman readership in mind. Although this sequence includes events that would certainly be strikingly unexpected if they actually occurred, they are not really miracles in the sense defined above. In function they are closer to dreams, prophecies and omens.

Finally, one should take a brief look at Josephus’s other two works, the Life and Against Apion. There is virtually nothing in the Life that could count as a miracle, although there are a few references to the workings of God’s providence (such as the sighting of a ship at Life 15). The closest one gets to anything numinous is a marvellous (θαυμάσιον) vision Josephus claims to have had in a dream when on the point of quitting Galilee (Life 208).

The nearest Against Apion comes to relating a miracle is the story in which Ptolemy Physcon arranges to have the Alexandrian Jews trampled by elephants, though, due to God’s intervention, the beasts trample Physcon’s friends instead, following which he sees a ‘terrible apparition (aspectum terribilem)’ (Apion 2.53–56). A similar story, in which Ptolemy Philopator is the villain, is told in more detail at 3 Maccabees 3–6.

Apart from that, Against Apion displays virtually no interest in miracle. At first sight this is odd, since the work is concerned to rebut calumnies against Moses, and in the Antiquities it is in relation to Moses that the greatest concentration of miracles is to be found. Yet in Against Apion Josephus makes no appeal to divine vindication of Moses by miracle. He seems more concerned to demonstrate the manifest absurdity of hostile slanders. His argument has a humanistic strain, stressing how wise Moses was in framing his laws in comparison with those of the Greeks, and how virtuous the Jews were in keeping them. Here, the Jews are praised for having a worthy idea of God rather than for being the recipients of God’s revelation or God’s saving acts.

This approach was probably dictated by the nature of the charges Josephus was rebutting or of the audience he was addressing. For example, at Apion 2.125–32 Josephus is obliged to argue that the Jews’ misfortunes do not discredit them, and this would make it awkward also to argue that they were vindicated by God’s interventions on their behalf. But whatever the reason, the impression given is that the theology of Against Apion has no need of miracles: what really matters is keeping the law, and Jewish devotion to the law is proved in adversity; if God grants political power to Rome that is secondary. Given that Josephus can take such a view, one wonders how important the miraculous can have been to him. This in turn raises the question of the function of the miraculous in Josephus’s works, which will form the topic of the next section.

The Function of the Miraculous in Josephus

This section is concerned with the function of the miraculous at two levels. First it will elucidate what types of function particular miracles perform within Josephus’s narratives; and second, it will investigate the general function of the miraculous for Josephus’s narrative discourse. These two tasks complete, it will finally discuss how seriously Josephus took the miracle stories he relates.

Josephus’s narratives include punishment, deliverance, provision and authentication miracles, as well as some minor categories. Of these, the punishment miracles form the largest class. The offences punished fall into four main groups: general immorality (e.g. Ant. 1.203 and War 4.483–84; Ant. 1.75–103, 163, 207–12, 14.28, 15.243; 17.168), specific disobedience (e.g. Ant. 8.240, 389), rebellion (e.g. Ant. 4.51, 56) and abuse of the holy (e.g. Ant. 3.209; 6.3, 5, 16; 7.82; 9.225; 12.413).

In some cases the offence may belong in more than one category. Thus, for example, at Ant. 4.131–40 Josephus narrates the seduction of the young Israelite males by the Midianite women. At 4.141–49 Zambrias (Zimri) makes a speech justifying this conduct by complaining that Moses’ legislation is too restrictive of liberty. In what follows Phinees (Phineas) and his companions despatch Zimri and many of his followers with the sword (4.150–54), but their slaughter is evidently incomplete, for God destroys the rest of them with a pestilence, together with those of their relatives who incited them (4.155). Here the offence punished combines rebellion, immorality and disobedience. Zimri furnishes a distant prototype to a much later rebel and excessive lover of liberty, Judas the Galilean (18.4, 23) of whom Josephus censoriously remarks, ‘Here is a lesson that an innovation and reform in ancestral traditions weighs heavily in the scale in leading to the destruction of the congregation of the people’ (18.9), a lesson the story of Zimri anticipates.

In other cases the punishment inflicted on one party is the shadow side of the deliverance afforded to another (normally Israel). Thus, for example, the plagues inflicted on Egypt (Ant. 2.294–314) both effect Israel’s escape and punish Pharaoh for his stubbornness. Similarly the plague inflicted on the Assyrian army is both a deliverance for Jerusalem in response to Hezekiah’s prayer and a punishment for Sennacherib’s arrogance and duplicity (Ant. 10.1–21; War 5.387–88, 404–408). On a lesser scale, the fire that comes down from heaven at Elijah’s behest both protects the prophet and punishes those who are trying to arrest him (Ant. 9.23).

The types of punishment inflicted range from the mundane to the spectacular. There is nothing obviously miraculous about being killed by lions or succumbing to pestilence except where such misfortunes fall swiftly and precisely on those whose misconduct has merited them. The dysentery and mice that attend possession of the ark seem so doubtfully miraculous to the Philistine rulers that they put the matter to the test by seeing where oxen left to their own devices will convey the troublesome object (Ant. 6.7–12). On the other hand, some of the punishments Josephus narrates are so spectacular that their miraculous nature is unmistakable. The earthquake that swallows Dathan and his company (Ant. 4.51–53) is one obvious example; not only is the event itself quite extraordinary (since the ground swallows only the guilty persons and then closes up leaving no trace) but it is preceded by Moses’ prayer that God should remove these men by such extraordinary means that his direct intervention cannot possibly be in doubt (4.47–49). Equally spectacular are the events that follow King Uzziah’s attempt to offer a sacrifice (9.225). Not only is the king smitten with leprosy, as at 2 Chron. 26:19, but there is a great earthquake which splits the temple and causes half the western hill to break off and roll round to the eastern hill (cf. Zech. 14:5).

There is no obvious correlation between either the severity or the miraculousness of these punishments and the heinousness of the crime committed; it seems rather that Josephus simply took over what lay in his source together with the occasional legendary embellishment. There is, perhaps, a sub-text in relation to the punishments inflicted on the rebellious Israelites in the wilderness period: the true followers of Moses in Josephus’s day are those who get quietly on with the business of obeying the law, whereas the hotheads who brought disaster by stirring up rebellion against Rome are the heirs of Dathan, Korah and Zimri. But what all these punishment miracles have in common is that they illustrate the negative aspect of the moral lesson Josephus spells out in the introduction to the Antiquities:

the main lesson to be learnt from this history by any who care to peruse it is that men who conform to the will of God, and do not venture to transgress laws that have been excellently laid down, prosper in all things beyond belief, and for their reward are offered by God felicity; whereas, in proportion as they depart from the strict observance of these laws, things (else) practicable become impracticable, and whatever imaginary good thing they strive to do ends in irretrievable disasters (1.14).

Having seen how punishment miracles illustrate the theme of ‘irretrievable disasters’, we must now look at those divine interventions that cause their recipients to ‘prosper … beyond belief,. Here deliverance miracles form the most significant class. These broadly fall into two types, miracles of national and of personal deliverance. The classic miracle of national deliverance in Josephus as for Jewish tradition generally is the crossing of the Red Sea (Ant. 2.329–49). In this instance Josephus leaves no doubt that there was a ‘clear manifestation of God’; whereas according to Exod. 14:21, ‘the Lord drove the sea back by a strong east wind all night, and made the sea dry land’, in Josephus’s account Moses strikes the sea with his staff and it recoils at once, a miracle if ever there was one. Elsewhere, the miracles of national deliverance tend to take the form of God’s help in securing a military victory. At Ant. 5.27 the wall of Jericho collapses ‘without either engine or force of any other kind having been applied to it by the Hebrews’. At 5.60 God helps Joshua in his campaign against the Canaanites with ‘thunder-claps, the discharge of thunder-bolts and the descent of hail of more than ordinary magnitude’. The idea that God fights for Israel with special weather conditions is a recurring motif; Deborah’s victory over Sisera is similarly aided by a great tempest (5.205), while Samuel’s defeat of the Philistines is facilitated by divinely sent earthquake, thunder and lightning (6.27).

It is noteworthy that Josephus represents God as helping the Romans against the Jews in the same way. For example, in describing the fall of Gamala, Josephus states that ‘to seal their [sc. the Jews’] ruin, a storm miraculously arose which, blowing full in their faces, carried against them the arrows of the Romans and checked and deflected their own’ (War 4.76; cf Ant. 5.205). The fall of Masada is likewise facilitated by the veering of the wind (War 7.318). Again, the hurricane, thunder and earthquake that accompany the Idumaeans’ arrival outside the walls of Jerusalem are generally taken as portents (War 4.286–87) but in fact allow the Zealots to admit the Idumaeans under cover of the din (4.298–99), thereby contributing to the downfall of the city. Here, one feels, Josephus is deliberately turning Israel’s sacred traditions on their head, in order to undermine their appropriation by the rebelliously minded.

The story of Elisha’s capture of the Syrian army (Ant. 9.51–59) is different from these weather-miracles, for this is an account not of a crushing victory aided by thunder-bolts, but of an effortless capture effected through divinely induced blindness. This particular story lies on the border between miracles of national and personal deliverance, for although the outcome is a victory of sorts for Israel, the main effect of the miracle is that Elisha evades capture.

It is the miracles of personal deliverance that predominate from then on (Jerusalem’s deliverance from the Assyrian army being the exception). These include the stilling of the storm that saves the other people on the ship when Jonah is cast into the sea (Ant. 9.212), Jonah’s own safe arrival on land (9.213), Daniel’s three friends emerging unscathed from the furnace (10.214), and Daniel surviving his incarceration in the lion’s den (10.258–62). This shift largely reflects what Josephus found in the scriptures; Josephus’s subsequent treatment of post-biblical history shows he did not think that God had withdrawn his interventions from the political sphere altogether.

The second main class of beneficent miracles are those of provision, which again may be aimed at large groups (such as Israel in the wilderness) or at individuals. These usually take the form of the miraculous supply of food or water in times of need. The principal examples from the wilderness period are the miraculous provision of water (Ant. 3.6–8, 37–38), quails (3.23–25) and manna (3.26–30). Subsequently food tends to be miraculously produced only for individuals, namely Elijah, Elisha and their associates (8.319, 323, 349; 9.48–50), while Samson is granted his own private spring (5.303). Water, however, continues also to be provided in large quantities, either by means of unexpected rain (8.346; 14.22, 390; 18.285) or without it (9.36–37). At War 5.409–10 the tradition of divine aid to Israel is once again inverted when the springs flow more copiously for Titus than for the Jews he is besieging.

Miracles of authentication may indirectly benefit those authenticated, and may have a variety of effects on those at whom they are targeted. They may simply provide a spectacle, as in the case of the signs Moses performs before Pharaoh (Ant. 2.284) or they may cause harm, such as the temporary paralysis of Jeroboam’s hand (8.233). This class of miracle has already been dealt with in the discussion of σημεῖα above. Perhaps the budding of Aaron’s rod (Ant. 4.64–65) falls into the same category, in that it confirms God’s choice of Aaron as high priest (though Josephus stresses its role in quelling yet another sedition).

A smaller class of miracles demonstrates God’s presence in some way, either through theophanies as at Sinai (Ant. 3.80 and its pale echo at 8.352), or on a lesser scale, such as the burning bush (Ant. 2.265–66) or the cloud over the tabernacle (3.202–203; cf. 8.106). Another small class of miracles signifies not so much God’s presence as his acceptance of a variety of sacrifices by miraculously consuming them with fire (Ant. 3.207; 5.284; 8.118). The fire that consumes Elijah’s sacrifice on Mount Carmel (8.337–42) is a special case. It does, of course, imply God’s acceptance of this sacrifice, just as it does authenticate Elijah as his true prophet, but as Josephus tells the story its main point is to demonstrate that the God of Israel is ‘the Almighty and only true God, while the others were mere names invented by unworthy and senseless opinion’ (8.343).

There are one or two miraculous happenings that do not fit neatly into any of these categories. The writing on the wall at Belshazzar’s feast (Ant. 10.233) is more a warning of judgment to come than a punishment miracle proper. The same might be said of the repeated prostration of the Philistine god Dagon in the presence of the ark (6.1–2). Balaam’s talking ass (4.109) is a curious way of presaging the appearance of an angel. All these, of course, are what Josephus found in the biblical record, though the same cannot be said of the curiously cooperative rainfall during the building of Herod’s temple (15.425).

Josephus narrates few miracles of healing. Perhaps the only story that strictly falls within this category is that of Hezekiah’s recovery from illness (Ant. 10.29) where, however, the emphasis falls on the sign of the retreating shadow. Elijah’s resuscitation of the widow’s son (8.325–27) is perhaps another instance, especially since Josephus hints at one point that the boy was not really dead (δόξαι νεκρόν, 8.325). The revival of the corpse thrown into Elisha’s tomb (9.183) comes closer to magic.

This material does not totally exhaust Josephus’s interest in healing. At Ant. 8.46–49 he describes how a Jew named Eleazar performed exorcisms in front of Vespasian by means of a ring containing ‘one of the roots prescribed by Solomon’. This is given by way of illustrating that ‘God granted him [sc. Solomon] knowledge of the art used against demons for the benefit and healing of men’. Perhaps the prescribed root was the one from Baaras whose curious properties, which include the power to expel demons, Josephus describes at War 7.180–85. The Jewish War also contains an extended description of the Essenes, in the course of which Josephus states

They display an extraordinary interest in the writings of the ancients, singling out in particular those which make for the welfare of soul and body; with the help of these, and with a view to the treatment of diseases, they make investigations into medicinal roots and the properties of stones (War 2.136).

But this additional material appears to lie more in the realm of magic or medicine than of miracle; healing miracles in Josephus remain rare. This is not simply due to the scarcity of this type of miracle in the Old Testament, for, as we have already observed, it is precisely this type of miracle that Josephus tends to omit: for example, the bronze serpent that saves from snake-bite (Num. 21:4–9), the cures from leprosy of Miriam (Num. 12:9–16) and Naaman (2 Kgs 5), and the resuscitation of the Shunammite woman’s son (2 Kgs 4:11–37). We have already reviewed the Elisha material, and Josephus probably had apologetic reasons for not mentioning leprosy in connexion with the exodus (namely Manetho’s slander that it resulted from the expulsion of Egyptian lepers, Apion 1.228–87). It nevertheless appears that Josephus had little interest in miracles of healing.29

In discussing the functions of the various types of miracle Josephus relates, I have already begun to adumbrate the overall function of the miraculous in his narrative discourse. Of course, the Jewish scriptures contained a considerable amount of miraculous material, so that Josephus could hardly have excluded it all without making a nonsense of his claim to be setting forth the precise details of scripture, ‘neither adding nor omitting anything’ (Ant. 1.17). But there is no indication that his adoption of this miraculous material is reluctant. On the contrary, he makes good use of it to illustrate the main moral lesson of his work (Ant. 1.14—see above). For Josephus miracle functions as the most striking manifestation of God’s πρόνοια.

This can be illustrated from several passages, many of which we have met before. It will be recalled that when Moses shows his three signs to the Egyptian king, Pharaoh is far from impressed, since he has magicians who can perform similar tricks. Moses thus undertakes to ‘show that it is from no witchcraft or deception of true judgement, but from God’s providence and power (πρόνοιαν καὶ δύναμιν) that my miracles proceed’ (Ant. 2.286). This contrast is instructive: a true sign, as opposed to a ‘deception of true judgement’, is something that issues from God’s providence.

Pharaoh remains unimpressed, and thereby brings the plagues upon Egypt. After the first three he relents to the extent of allowing the Hebrew men to depart, but insists that their wives and children remain behind as hostages. In Josephus’s view, this was an imposition ‘upon [God’s] providence, as though it were Moses and not He who was punishing Egypt on the Hebrews’ behalf, (2.302). This shows that God’s providence may manifest itself in punishment as readily as in reward or care; it further illustrates how Josephus uses the word πρόνοια to indicate God’s control of events.

This is most dramatically illustrated in what Moses says before the crossing of the Red Sea. As in the Biblical account, the escaping Hebrews are hemmed in, believing themselves to be in a desperate plight from which no escape is possible. Moses’ prayer emphasizes that, humanly speaking, escape is indeed impossible (Ant. 2.335), so that they have no recourse but to look to God’s providence (2.336); yet since the sea and the mountains that hem them in are God’s, his almighty power will surely be able to open a way through them, or even, perhaps, through the air (2.337). In the speech that precedes this prayer Moses rebukes the frightened Hebrews for despairing of God’s providence given what he has already achieved for them (2.330), and then goes on to tell them that God has led them into their present dire strait so that ‘He may display both His own power and His tender care (πρόνοιαν) for you. For it is not in trivial circumstances that the Deity lends His own aid to whom He favours, but where He sees men have lost all hope of ameliorating their lot’ (2.332). Two points in particular are worth noting here. The first is that Josephus appears to regard a miraculous deliverance as the most proper demonstration of God’s providential care (‘not in trivial circumstances’). The second is that the idea of God’s providential care for those ‘whom He favours’ has largely replaced the biblical idea of covenant.

The fourth passage that demonstrates the close link between miracle and providence is another story we have had occasion to look at more than once. When Moses asks that Abiram and Dathan be removed ‘in no common manner out of existence’, this is in order to ‘Prove now once again that all is directed by thy providence, that nothing befalls fortuitously (αὑτομάτως), but it is thy will that overrules and brings everything to its end’ (4.47–48). This is a considerable elaboration on what Moses says at Num. 16:28–30, and may thus be regarded as a statement of Josephus’s own position. Here providence is effectively equated with God’s will (‘all is directed by thy providence … thy will … brings everything to its end’) and contrasted with events happening ‘fortuitously’.

This is not the only place Josephus makes this contrast. At 10.277–81 he argues from the accuracy of Daniel’s prophecy:

how mistaken are the Epicureans, who exclude Providence (πρόνοιαν ἐκβάλλουσι) from human life and refuse to believe that God governs its affairs or that the universe is directed by a blessed and immortal Being to the end that the whole of it may endure, but say that the world runs by its own movement (αὑτομάτως φέρεσθαι) without knowing a guide or another’s care.

I have not included prophecy as a type of miracle, but Josephus may not make such a sharp distinction; in any case, he makes it abundantly clear that Daniel’s prophecies come from God (10.277). In a similar way the justification Josephus gives for narrating Archelaus’s and Glaphrya’s dreams (17.345–54) is that ‘they provide instances of something bearing on the immortality of the soul and of the way in which God’s providence (προμηθείᾳ) embraces the affairs of man’. This further illustrates the connexion in Josephus’s thought between God’s providence and unusual phenomena, as well as showing that the connexion between the two is not confined to the Mosaic period.

A few further examples may serve to establish this latter point. When Daniel’s three friends are thrown into the fiery furnace, it is by divine providence that they miraculously escape death (θείᾳ σώζονται προνοίᾳ καὶ παραδόξως διαφεύγουσι τὸν θάνατον, 10.214). Daniel himself emerges unscathed from the lions’ den, and when his enemies refuse to believe that this ‘was through the Deity and His providence (διὰ μὲν τὸ θεῖον καὶ τὴν τούτου πρόνοιαν)’, Darius first throws a large quantity of food to the lions so that they may eat their fill, and then throws Daniel’s enemies in after them; the lions at once devour these men, thereby convincing Darius that it was God and not the lions’ satiety that had saved Daniel (10.260–62). On a less spectacular level, it is also ὡς ἐκ θεοῦ προνοίας that rain arrives just in time to prevent Herod’s brother Joseph from having to flee Masada while Antigonus is besieging it (14.391), and κατὰ θεοῦ πρόνοιαν that Herod escapes assassination in his bath (14.462).

To demonstrate the intimate connexion between miracle and providence throughout Josephus, or at least, throughout the Antiquities, one must do more than instance occurrences of the word πρόνοια. One must also show that the idea is present even when the word is not. But this has already been achieved. The types of miracle on which Josephus concentrates, punishments, rewards, and saving acts, are precisely those in which God’s providence (understood as his effecting ‘proper retribution for good and evil within history’) is most clearly demonstrated. This conclusion is further supported by the type of material Josephus chooses to include and omit in the Elisha cycle.

Thus, at least in the Antiquities, Josephus is engaged in a kind of narrative theology in which πρόνοια, frequently demonstrated by miracles, plays a key role. Of course πρόνοια is a characteristically Stoic word, but this does not mean that Josephus is using it in a characteristically Stoic way. Rather he seems to have picked it up from the popular miracle vocabulary of his time as a convenient term in which to express his understanding of the Jewish tradition in terms comprehensible to Graeco Roman readers.

From what has been said so far, the question of how seriously Josephus took the miracles he narrated virtually answers itself. For Josephus, miracles are the clearest manifestations of God’s providence, which in turn is one of his key theological categories. When it comes to choosing between a providential and a merely fortuitous explanation, Josephus may courteously invite his readers to decide for themselves, but by then he has already made it perfectly plain where he himself stands (e.g. Ant. 2.347–48; 17.354).

This conclusion is by no means undermined by Josephus’s occasional remarks to the effect that he has been constrained to record what is set forth in the scriptures (e.g. Ant. 2.347; 3.81; 10.218, 281). Far from being an attempt to shrug off responsibility for fantastic stories on to the sources he is paraphrasing, such statements must be read as an appeal to the authority of scripture in their support. For in the proem to the Antiquities Josephus stresses that Moses, his prime authority, held a view of God that was free from ‘unseemly mythology’ even though in writing about events so long ago he might have invented fictions (Ant. 1.15–16). Besides, it is implausible in the extreme to suppose that Josephus means to impugn the authority of scripture (Apion 1.42–43), or, for example, that his use of the formula at Ant. 3.81 could be intended to cast doubt on the law-giving at Sinai.

Neither is this conclusion seriously challenged by the fact that Josephus sometimes rationalizes the miraculous. For example, in relating the provision of quails in the wilderness, Josephus explains that this was a species of bird abundant in the Arabian gulf, and that they settled in the Hebrews’ camp because they were wearied from their flight (Ant. 3.25), whereas the biblical account merely states that ‘In the evening quails came up and covered the camp’ (Exod. 16:13). Nevertheless, not only is Josephus’s ‘rationalization’ here a perfectly reasonable expansion of the biblical account, but Josephus also makes it perfectly clear that God was behind this event:

God thereon promised to take care of them … Having received this response from God, Moses descended to the multitude; and … told them that he had come to bring them from God deliverance from their present straits … And they, collecting them as the food devised for them by God, assuaged their hunger; while Moses addressed his thankful prayers to God for sending succour so prompt and in keeping with His promise (3.24–25).

In this instance the ‘rationalistic’ additions may explain the miracle, but they do not explain it away.

In some instances it is not so clear how ‘rationalistic’ the proposed explanation is. In his account of the purification of the waters of Marah, Josephus adds the non-scriptural details that Moses had men draw off the undrinkable surface water, and that the remainder was purified by incessant blows (3.8). It is, however, less than clear how these operations would naturally produce the desired result. In any case Josephus again emphasizes God’s role, stating that ‘God had lent an ear to his prayers and had promised to render the water such as they desired, provided that they executed His orders with no remissness, but with alacrity’ (3.7). This insistence on instant obedience shows that this is no straightforward rationalization if by ‘rationalization’ one means a natural explanation that is intended to exclude divine intervention.

Moreover, there are many occasions on which Josephus either tells the story much as it appears in the Bible or else heightens the miraculous. We have already noted the latter phenomenon in connexion with the miracles of the Red Sea crossing and Daniel in the lions’ den, so the evidence need not be repeated here.

There is thus little to suggest that Josephus’s ‘rationalizations’ indicate any serious doubts about the veracity of the miracle stories he narrates. In all probability they are simply concessions to his Hellenistic readership. George MacRae strikes the right balance when he remarks

As a Jew, Josephus does not balk at accepting the miraculous wherever he encounters it because what he sees in it are God’s πρόνοια and δύναμις. But as a Hellenist, he does not hesitate to offer a pseudo-scientific or pseudo-philosophical explanation as well wherever one comes easily to mind.


In many respects the discussion of this chapter has been along the lines laid down by previous scholars such as Gerhard Delling, George MacRae, Otto Betz, Harold Attridge and Rebecca Gray. In calling upon a wider range of evidence this discussion has, however, attempted to refine the understanding of Josephus’s miracle vocabulary. It has also gone beyond previous studies in defining Josephus’s concept of miracle and discussing how he distributes his miraculous material throughout his writings. It has further enumerated the specific types of purpose for which Josephus regards miracles as being performed, which has highlighted how, unlike the Evangelists, he was particularly interested in miracles of punishment and national deliverance in war. Finally, it has corrected, or at least qualified, the impression given in some previous writers that Josephus tends to rationalize miracles.

It is of some significance for this study that Josephus did believe in miracles and regarded them as being of some importance. It is also interesting that, in common with the Evangelists, he saw that purported miracles could be ambiguous. But only in miracles of provision do Josephus and the Gospels really coincide in the type of miracle narrated, though on these we must reserve judgment until we have seen how similar miracles are treated elsewhere. Particularly noteworthy is Josephus’s comparative lack of interest in miracles of healing.

We shall return to particular aspects of Josephus’s writings in later chapters, but the next task is to examine the other first-century Jewish author whose surviving writings form a substantial corpus: Philo.

Chapter 3

Miracle in Philo


It is debatable how far Philo constitutes a variety of Judaism all by himself, or how far he is the representative of an Alexandrian tradition. Philo’s peculiar philosophical concerns are more apparent in his allegorical writings than in his literal exposition. The latter may be aimed either at pagans, or at Jews whom Philo feels need instructing in their ancestral religion. Either way, these writings are likely to be representative of a broader segment of Jewish opinion than Philo’s more esoteric works (this is also suggested by occasional similarities in the way both Philo and rabbinic midrash embroider the biblical narrative, which will be noted as occasion arises). They may thus reflect views sufficiently widespread in Hellenistic Judaism to be worth comparing with the Gospels.

A further reason for concentrating more on Philo’s exposition than his allegory is that in the former Philo treats miracles as historical events, whereas in the latter their miraculous nature is dissolved in the allegorical exegesis Philo foists upon them. For our purposes, Philo the re-teller of biblical stories is far more useful than Philo the allegorical exegete.

The Concept and Language of the Miraculous in Philo

In the previous chapter we saw that for Josephus, a miracle is a strikingly unexpected event that believers take to be an act of God. This definition would not work so well for Philo, for whom miracles are not just strikingly unexpected but humanly impossible. There is less of a tendency in Philo for the miraculous to shade off into the merely providential, and more of a tendency to demarcate miracles as extraordinary divine interventions. Along with this is a reduced sense of the epistemological ambiguity of miracles; whereas Josephus recognizes that one person’s miracle might be another’s magical trickery, Philo regards miracles as objective divine interventions that anyone can recognize as such. For Philo, a miracle may thus be defined as a special act of God (possibly through an intermediary) that accomplishes something otherwise impossible.

Philo comes closest to making such a definition explicit at Quaest. in Gen. 4.51, where in commenting on the destruction of Sodom and Gomorrah he remarks, ‘And this marvellous activity (ἡ τεθαυματουργημένη πρᾶξις) shows, not the established habit of the elements, but a certain autocratic and arbitrary power which transforms the elements of all things as it chooses.’ It is also implied in Philo’s frequent insistence that all things are possible to God, even those that are otherwise impossible. Thus, for example, when the Israelites are trapped between the Red Sea and the pursuing Egyptians Philo has Moses assure them, ‘When God gives help He needs no armament. It is His special property to find a way where no way is. What is impossible to all created being is possible to Him only, ready to His hand (τὰ ἀδύνατα μαντὶ γενητῷ πόνῳ δυνατὰ καὶ κατᾶ χειρός)’ (Vit. Mos. 1.174; cf Abr. 112, 175; Spec Leg 1.282; 4.127; Op. Mund. 46; Virt. 26; Somn. 1.87; Quaest. in Gen. 2.47; 3.56). In Philo’s view God is able to work miracles by virtue of being the Creator. Like Josephus, Philo is aware that some of his readers may balk at the miraculous events he describes, but whereas Josephus throws a sop to the sceptic by remarking that on such matters everyone must decide for themselves, Philo roundly declares

If anyone disbelieves these things, he neither knows God nor has ever sought to know Him; for if he did he would at once have perceived—aye, perceived with a firm apprehension—that these extraordinary and seemingly incredible events (τὰ παράδοξα δὴ ταῦτα καὶ παράλογα) are but child’s-play to God (Vit. Mos. 1.212).

The reason for this is that the universe with its heavenly bodies moving regularly, the vast variety of life on earth, rivers, fountains, air, the yearly seasons ‘and other beauties innumerable’ are really far more marvellous:

But these things, though truly marvellous (ἀλήθειαν ὄντα θαυμάσια), are held in little account because they are familiar. Not so with the unfamiliar; though they be but small matters, we give way before what appears so strange, and, drawn by their novelty, regard them with amazement (Vit. Mos. 1.213).

Philo does not mean that the miracles he describes pale into insignificance beside the great miracle of creation, but is rather arguing a fortiori that a God capable of working the grand miracle of creation is capable of working any other miracle attributed to him.

At Vit. Mos. 1.212 Philo asserts three things that are characteristic of his view of miracles: first, they are events that objectively occur; secondly they are events that are sharply demarcated from the normal run of things as being παράδοξα … παράλογα; and thirdly they are manifestly acts of God, for only God could perform them. Implicit in this is Philo’s denial of any epistemological ambiguity in the miraculous.

To demonstrate these points more widely from Philo’s writings one may start by showing how frequently he stresses the extraordinary and seemingly impossible nature of the miracles he describes. When Moses receives his commission from God he is promised ‘three signs which no man has ever before seen or heard of’ (Vit. Mos. 1.76). Pharaoh remains stubborn despite being shown these signs, and the ten plagues that follow are described in terms that both heighten and emphasize their miraculous nature. Philo starts by informing his readers that ‘The chastisement was different from the usual kind, for the elements of the universe—earth, fire, air and water—carried out the assault’ (Vit. Mos. 1.96). Just how different becomes apparent when Aaron initiates the first plague. It is not just the Nile that turns to blood, but every last drop of water in the kingdom, ‘the lakes, canals, springs, wells and fountains and all the existing water-supply of Egypt’ (1.99); moreover, this all happens at once. When Aaron then summons a plague of frogs, these are not merely numerous enough to be irksome, but were ‘so numerous that not only the market-places and all the open spaces, but all the farm-buildings as well, and houses and temples and every place, public or private, was filled with them’ (103). Before describing the hailstorm, Philo explains at some length why rain is an exceptional event in the Egyptian climate (114–17). The hailstorm, accompanied by thunder and lightning, is then quite clearly a miraculous event, a fact Philo further underlines by remarking that the thunderbolts ‘provided a most marvellous (τερατωδεστάτην) spectacle, for they ran through the hail, their natural antagonist, and yet did not melt it nor were quenched by it’ (118). Likewise the darkness that suddenly fell over the land in daytime could be of no ordinary nature (123). Although Philo suggests that the darkness may have been due to a solar eclipse or unusually heavy cloud cover, these themselves would have been miraculous since the darkness lasts for three full days, and Philo has already intimated that storm-clouds do not occur in the Egyptian climate. In any case, Philo excludes any natural explanation by remarking that though the Egyptians were subject to this impenetrable darkness, the Hebrews, who were living among them, ‘lived in clear radiance with the light of day shining upon them’ (145). Whereas in the biblical account this could be explained on the basis that the Hebrews were living in a separate area (Exod. 8:22; 9:26), in Philo’s version ‘they dwelt in the same cities and villages and houses’ as the Egyptians (143). That Philo draws attention to this feature in connexion with all the plagues shows that he is not concerned merely to show the gravity of the punishments that fell on the Egyptians, but also to stress how miraculous they were:

And the strangest thing of all (ὅ δὴ καὶ παραδοξότατον) was that the same elements in the same place and at the same time brought destruction to one people and safety to the other. The river changed to blood, but not for the Hebrews; for, when they wished to draw from it, it turned into good drinking-water. The frog tribe crept from the water on to the land, and filled the market-places, the farm buildings and houses, but held aloof from the Hebrews alone, as though it knew how to distinguish who should be punished and who should not (Vit. Mos. 1.143–44).

This stress does not stop here. Referring to the pillar of fire and cloud that accompanied the fleeing Israelites, Philo remarks that as they set out for the Red Sea, ‘It was then, we are told, that there occurred a prodigy (τεράστιον), a mighty work of nature (μεγαλούργημα τῆς φύσεως), the like of which none can remember to have seen in the past’ (165). The parting of the waters broadly follows the biblical account, but nevertheless leaves no doubt that the event was absolutely miraculous: ‘Moses now, at God’s command, smote the sea with his staff, and as he did so it broke and parted into two. Of the waters thus divided, one part rose up to a vast height, where the break was made, and stood quite firmly, motionless and still like a wall’ (177). Philo explicitly describes the subsequent movement of the pillar of cloud to act as the Israelite’s rear-guard as ‘a most extraordinary sign (σημεῖον … τερατωδέστατον)’ (178). Elsewhere he describes the whole Red Sea crossing as ‘This wonderful sight and experience, an act transcending word and thought and hope’ (Vit. Cont. 87), and the pathway that appeared through the sea as ‘a road of miracle (μεγαλουργηθείσης ὁδου)’ (Vit. Mos. 2.253). In De vita Mosis 1 he further remarks of the destruction of the pursuing Egyptians that ‘This great and marvellous work (τὸ μέγα τοῦτο καὶ θαυμαστὸν ἔργον) struck the Hebrews with amazement’ (Vit. Mos. 1.180).

The same emphasis continues in Philo’s account of the wilderness wanderings. Moses is indignant that the Hebrews continue to grumble ‘after experiencing strange events outside the customary without number’ (1.196). God, nevertheless, moved by the plight of the hungry mob ‘devised new and strange forms of benefaction, that by clearer manifestations they might now be schooled not to shew bitter resentment’ (199). These take the form of a strange, unprecedented rain (of manna), an incredible sight ‘which no man ever saw before’ (200). But that is not all: ‘This work of God was strange enough even if it had stopped at this point, but actually there were other facts still stronger enhancing its marvels (ἐθαυματουργεῖτο)’ (203), namely that the manna could not be kept for future use except when collected on the day before the sabbath. As with the plagues that selectively afflicted the Egyptians but not the Hebrews, this torah-supporting behaviour of the manna cannot be accounted for by merely ‘natural’ causes. Thus Philo subsequently notes that when manna was made available for the sabbath by the double ration collectable the day before, ‘the greatness of the wonder (τέραστιον) was shown not only by the double supply of food and its remaining sound contrary to the usual happening, but by the combination of both these things occurring on the sixth day’ (Vit. Mos. 2.266). But even this is not the greatest thing according to Philo, since in his view:

There is something still more wonderful (θαυμασιώτερον) to be told. During all that long period of forty years in which they journeyed, the food required was supplied according to the rules just mentioned, like rations measured out to provide the allotment needed for each (Vit. Mos. 1.206).

It is in this connexion that Philo emphasizes Moses had good cause to be indignant when the people initially ignored his instructions on collecting the manna ‘after witnessing wonders so many and so great, impossibilities (ἀδύνατα) no doubt as judged by what to outward appearance is credible and reasonable but easily accomplished by the dispensations of God’s providence’ (Vit. Mos. 2.261; cf. Vit. Mos. 1.212).

There are many other examples of language emphasizing the sheer extraordinariness of miraculous events, for instance in the quelling of Korah’s rebellion (Vit. Mos. 2.281–87), the fire from heaven that consumed the first priestly sacrifices (Vit. Mos. 2.154), the fact that Abraham’s three angelic visitors took on human form and were able to give the appearance of eating and drinking (Abr. 118), or that the rain of fire on Sodom was an ‘unprecedented calamity of God’s mighty working’ (Abr. 142), the statement that the Israelites ought to have been awe-struck by the marvel that provided them with quails in the desert (Spec. Leg. 4.129), the fact that God chose a barren woman (Sarah) to be the mother of the race ‘in order that the seed of offspring may appear more wonderful and miraculous’ (Quaest. in Gen. 3.18), and the punishment of Lot’s wife, turned into a pillar of salt, as a warning ‘by producing even more wonderful miracles’ (Quaest. in Gen. 4.52). Although Philo frequently displays a fondness for exaggeration, the emphatic language in which he stresses the marvellous or unprecedented nature of miraculous events is nevertheless significant; and although most of this occurs in De vita Mosis, there are plenty of examples elsewhere as well to show that, for Philo, a miracle is an extraordinary event sharply distinguished from the normal run of things.

This is not to say that Philo would have defined a miracle as a violation of the laws of nature. Even though the phrase νόμος φύσεως (and its near equivalent, λόγος θύσεως) occurs in Philo’s writings, it does not bear the modern sense of ‘law of nature’. When Philo praises the Torah as being unique (in distinction from merely human law-codes) in reflecting the laws of nature (e.g. at Vit. Mos. 2.52; cf. Abr. 5–6) he does not mean that the Pentateuch is a physics textbook. The ‘laws of nature’ are rather an expression of the Creator’s will for his Creation (see, e.g., Poster. C. 185, where the ‘law of nature’ is equated with virtue). Philo admittedly has a strong sense of the regularity and law-like behaviour of the created order when it is behaving normally, but he is also clear that it is a created order, and not an autonomous system. What God can create, God can also make perform in unusual ways, for ‘God has subject to Him not one portion of the universe, but the whole world and its parts, to minister as slaves to their master for every service that He wills’ (Vit. Mos. 1.201).

Finally, although Philo does not equate God with Nature in the Stoic sense, he sometimes uses ‘Nature’ to mean ‘God’ in his own sense (e.g. Vit. Mos. 2.27; Jos. 38, 192). In performing a miracle Philo’s God does not so much intervene in nature as command its elements to work in unaccustomed ways, exchanging their wonted behaviour with that of other elements.

Along with Philo’s more sharply demarcated notion of miracle (compared with Josephus) goes a stronger notion of its evidential value. For Philo a miracle is not simply an objective event, it is objective proof of God’s special activity. Thus whereas Josephus’s Pharaoh goes on believing that Moses is merely an accomplished magician able to out-perform those of his own court, Philo’s Egyptians quickly recognize what they are really up against. At Vit. Mos. 1.76 God undertakes to give Moses ‘three signs which no man has ever before seen or heard of [which] will be sufficient lesson to convert’ those who disbelieve him. When these are subsequently performed before the king of Egypt, his magicians undertake to replicate them. They too throw down their rods, which duly turn into serpents. But Aaron’s serpent swallows each of theirs and then turns back into a rod, with the result that

By this time, the marvellous spectacle (ἡ μεγαλουργηθεῖσα ὄψις) had refuted the scepticism in every ill-disposed person’s soul, and they now regarded these events not as the works of human cunning or artifice fabricated to deceive, but as brought about by some diviner power to which every feat is easy. But, though they were compelled by the clear evidence of the facts to admit the truth … They did not show mercy to those who were unjustly enslaved, nor carry out the orders which had divine authority, since God had shown His will by the proofs of signs and wonders (σημείων καὶ τεράτων) (Vit. Mos. 1.94–95).

In Josephus’s account of the same events (Ant. 2.284–87) all Pharaoh sees is a competition between two groups of magicians, and he supposes that Moses is simply ‘trying to impose on him by juggleries and magic’. In contrast, Philo regards these signs as objective proofs of God’s action even for the most ill-disposed observer.

When Pharaoh’s stubbornness necessitates the ten plagues, Philo remarks in retrospect ‘never was judgement so clearly (ἐμθανῶς) passed on good and bad’ (Vit. Mos. 1.146). Subsequently, when the Israelites grumble at Moses in the wilderness, Moses is justifiably indignant at them, ‘For, after experiencing strange events outside the customary without number, they … should have put their trust in him of whose unfailing truthfulness they had received the clearest proofs’ (196). Again, after Moses has instructed the first priests in their duties, he and Aaron pray before the altar, whereupon a miraculous fire consumes the offering upon it, ‘thus giving, I hold, the clearest proof that none of these rites was without divine care and supervision’ (Vit. Mos. 2.154). When the people ignore Moses’ instructions on gathering manna, ‘the Father confirmed the utterance of the prophet with two most convincing proofs’ (Vit. Mos. 2.262). In response to Korah’s rebellion, Moses prays that his oracles should be confirmed by the rebels meeting a death ‘of a new and different kind’ so that his ‘truthfulness will be attested’ (Vit. Mos. 2.281). According to Philo, one of the reasons for giving the law in the desert was that the miraculous provision of food and drink there should show these laws to be the pronouncements of God (Dec. 16). These examples all show that for Philo a miracle is objective proof of divine activity.

It hardly needs demonstrating that for Philo a miracle is a divine action. This has already become apparent. Passages that explicitly assert the divine authorship of miracles include Vit. Mos. 1.71, 76, 95, 97, 107, 108, 112, 134, 174, 185, 199, 201–205, 209, 212; Vit. Mos. 2.154, 177–79, 213, 253, 261, 262, 266, 284; Op. Mund. 40; Abr. 96, 110–12, 137, 142; Dec. 15–17, 33; Spec. Leg. 2.199, 218; 4.127; Virt. 45–46; Det. Pot. Ins. 177; Congr. 173; Quaest. in Gen. 1.32; 2.28, 47; 3.18, 56; 4.17, 51. God’s role as BNP will in any case be demonstrated when we come to look at Moses’ (and Aaron’s) miracle-working role in the next section.

The foregoing discussion has surely justified the definition with which I began, namely that for Philo a miracle is a special act of God (possibly through an intermediary) that accomplishes something that would otherwise be impossible. The next task is to examine the language Philo uses in connexion with miracle.

Josephus proved unwilling to use θαῦμα and its compounds in connexion with miracle, reserving the word for ‘wonder’ or ‘marvel’ in a more general sense. No such reluctance is displayed by Philo, although he seldom uses the bare word θαῦμα as opposed to one of it longer compounds. Miracles are sometimes described as θαυμαστός (Vit. Mos. 1.180) or θαυμάσιος (Vit. Mos. 1.206; Quaest. in Gen. 1.86), but, as in Josephus, these words are more often used to describe something wonderful or marvellous in a non-miraculous sense (e.g. Hyp. 7.11, 17; Vit. Mos. 2.10, 17; Spec. Leg. 1.342; Praem. Poen. 115; Agr. 71; Migr. Abr. 33, 210; Quis. Her. 95; Mut. Norn. 264; Aet. Mund. 41; Omn. Prob. Lib. 63). The θαυμ- words Philo reserves for the miraculous are those formed from the passive of θαυματουργέω, either as a past participle (e.g. Vit. Cont. 85; Vit. Mos. 1.71; 2.179) or as a verb (e.g. Vit. Mos. 1.79, 203; Abr. 118; Dec. 33, 44; Spec. Leg. 2.218; Quaest. in Gen. 4.52) or as the noun θαυματουργήμα (Vit. Mos. 1.82). These word-forms all suggest a wonder or marvel that has been worked; they thus come closer to expressing our word ‘miracle’ than would θαῦμα and its cognate adjectives by themselves.

There are only two instances in Philo’s writing where θαυματουργέω is used in a sense other than that of working miracles. In the first of these, Mut. Nom. 162, the vine is described as ‘a wonderful piece of nature’s handiwork (ὑπὸ φύσεως τεθαυματούργηται)’. The second, Somn. 2.119 may not be not so much a genuine exception as a piece of sarcasm. In this passage Philo is describing the vaunting ambition of Xerxes in bridging the Hellespont, an act Philo describes as ‘creating a revolution in nature; for he converted two elements, earth into sea, and sea into earth, giving dry land to the ocean and ocean in exchange to the dry land’. This language of exchanging elements is close to the way Philo sometimes talks about miracle elsewhere. In the present context Philo goes on to say ‘And having played the conjurer (θαυματουργήσας) … [Xerxes] proceeded in the boldness of his schemes to mount to heaven also.’ Philo appears to be mocking Xerxes as a would-be god, so that the Loeb translator’s ‘conjurer’ could reasonably have been rendered ‘miracle-worker’; in performing such massive efforts to bridge the Hellespont and create an artificial sea in place of Mount Athos, Xerxes aspires to perform a ‘miracle’. That being so, it may be concluded that Philo reserves the word θαυματουργέω to the sense of ‘miracle-working’ with only one real exception (the ‘miraculous’ vine).

Philo’s second characteristic set of words for miracle are those formed from the passive of μεγαλουργέω, mostly past participles used as adjectives or cognate nouns meaning ‘mighty work’ (e.g. Vit. Mos. 1.66, 94, 165; Vit. Mos. 2.253, 251; Abr. 142; Spec. Leg. 2.188; 4.129; Leg. Gai. 85). This is not, however, a word that Philo reserves exclusively for miracles, since he also uses the word in the pejorative sense of ‘enormity’ (e.g. Spec. Leg. 2.170; 3.151; Conf. Ling. 157; Praem. Poen. 134; Flacc. 180; Leg. Gai. 213, 348; cf. μεγαλουργία at Somn. 2.117 and μεγαλουργός at Flacc. 73; Leg. Gai. 194). These are, however, two quite distinct usages, which do not merge into each other at any point (except, perhaps, where Philo is discussing Xerxes in Somn. 2).

Philo is also happy to use τέρας (e.g. Vit. Mos. 1.80, 90, 91, 95; Spec. Leg. 2.218; Agr. 96) or τεράστιον (e.g. Vit. Mos. 1.71, 165; 2.266; Abr. 118; Spec. Leg. 2.188; 4.129; Congr. 173) to describe a miracle (cf. Exod. 4:21; 7:3, 9; 11:10; 15:11 where lxx uses τέρας, usually but not always in conjunction with σημεῖον, to mean ‘miracle’). He also likes using the superlative of adjectives derived from τέρας (Vit. Mos. 1.118, 217; 2.154, 268; Abr. 118). He seldom uses these words outside miraculous contexts, though there are a few exceptions. Talking about the creation at Plant. 4, Philo asks, ‘How can it be other than a prodigy (τεράστιον) that the dissolving element should be held together by that which it dissolves, water by earth; and that on the coldest element the hottest should be seated unquenched, fire upon air?’ But perhaps Philo does regard these things as miraculous, even though they do not strike us as being so. Again, at Aet. Mund. 69, where Philo is talking about the eternity of the human race in the course of an argument for the eternity of the world, he remarks, ‘For the race remains for ever, though particular specimens perish, a marvel (τεράστιον) in very truth and the work of God.’ Τεράστιος is used of marvel in a pejorative sense at Quod. Omn. 5. At Spec. Leg. 3.45 τέρας is used to mean ‘monstrous hybrid’, and at Aet. Mund. 121 (in a quotation from Pindar), perhaps merely ‘marvel’. There is thus a strong tendency in Philo to reserve τέρας and τεράστιος to the senses ‘miracle’ and ‘miraculous’, but he appears less careful to do so with this word, which he probably borrowed from the Septuagint, than with the longer words of his own choosing.

Philo occasionally uses σημεῖον to mean ‘miraculous sign’. For example, he refers to the three ‘signs’ that God gives Moses at the burning bush (Vit. Mos. 1.76–77). Although this coincides with Josephus’s use of σημεῖον in his account of the same incident at Ant. 2.274, there is little to suggest that Philo is anticipating Josephus’s special usage. For one thing, Philo uses the words ἐθαυματουργεῖτο (Vit. Mos. 1.79) and τεράτων (80) to refer to the same three signs, which suggests he regards these terms as interchangeable. For another, in calling Moses’ authenticating miracles σημεία, Philo is probably echoing the lxx of Exod. 4:8–9. At Vit. Mos. 1.95 Philo refers to the first of these three demonstrations with the stock phrase ‘signs and wonders’, though in the same passage they are also called simply ‘wonders’ (τέρατα, 90) and ‘the marvellous spectacle’ (μεγαλουργηθεῖσα ὄψις, 94). Subsequent occurrences of the word ‘sign’ do not conform to any particular pattern. The movement of the pillar of cloud to act as the Israelites’ rearguard is described as ‘a most extraordinary sign’ (σημεῖον … τερατωδέστατον, 178). To obtain water in the desert Moses strikes a rock with ‘that sacred staff with which he had accomplished the signs in Egypt’ (Vit. Mos. 1.210). These signs presumably include at least some of the plagues (Vit. Mos. 1.120) if not the parting of the Red Sea (177), and so cannot be confined to the rod-snake demonstration. The phrase ‘signs and wonders’ is also used at Spec. Leg. 2.218, where, Philo states, in the course of a canticle in the feast of Baskets the celebrant recalls that God ‘confounded their assailants with signs and wonders and portents and all the other marvels that were wrought at that time’. It occurs again at Aet. Mund. 2 where Philo reflects that ‘perhaps God would not have refused to impart the knowledge of things heavenly through dreams or oracles or signs or wonders to souls thoroughly purged and bright and radiant’. Finally, at Det. Pot. Ins. 177 Philo comments on Gen. 4:15 (‘the Lord God set a sign [lxx: σημεῖον] upon Cain’) as follows:

and what the sign is, he has not pointed out, although he is in the habit of showing the nature of each object by means of a sign, as in the case of events in Egypt when he changed the rod into a serpent, and the hand of Moses into the form of snow, and the river into blood.

Admittedly, this once again connects σημεῖον to Moses’ three authenticating signs, but the connexion appears to take place through the words employed by the Septuagint rather than as a result of any particular nuance intended by Philo. Thus, although Philo can use σημεῖον to mean ‘miracle’, it is not one of his more important miracle words, neither is it one he restricts to any particular sense; it appears simply to be a word he has picked up from the biblical text and occasionally uses as a variation from his more characteristic miracle vocabulary.

Like Josephus, Philo also uses παράδοξος (e.g. Vit. Mos. 1.203, 212; 2.213) and παράλογος (e.g. Vit. Mos. 1.196, 212) in connexion with miracle. But whereas in Josephus these words, especially the former, almost become substantives meaning ‘miracle’, in Philo their use remains purely adjectival, to describe miraculous events as exceptional, surprising and contrary to expectation

Overall, then, we have found that although Philo has a variety of words for miracle, which he tends to use interchangeably, there are three in particular (θαυματουργ-, μεγαλουργ- and τέρας/τεράστιον) he uses most frequently and tends to reserve exclusively for miracle. This further supports the contention that Philo had a definite concept of miracle as a quite distinctive kind of event.

The Distribution of Miracles in Philo

Most of the examples employed above came from De vita Mosis or otherwise referred to the Mosaic epoch. This is hardly surprising given Philo’s concentration on the Pentateuch and the fact that most of the Pentateuchal miracles occur in connexion with the exodus, but it does mean that the present section will focus mainly on Philo’s treatment of the Mosaic miracles. This is no disadvantage if the aim is ultimately to compare Philo with the Gospels, since for Philo, Moses is the figure that comes closest to having the significance that Jesus had for the Evangelists. To be sure, one must not take Philo’s more extravagant language too literally, as, for example, where he appears to suggest that Moses was quasi-divine (Vit. Mos. 1.27) or even an incarnation of the Logos. The purpose of such language is more to glorify Judaism by glorifying its law-giver than to make an ontological claim about the person of Moses. Nonetheless, the fact remains that for Philo, Moses is the human being who approaches most closely to God, and through whom God’s will is most clearly revealed. In its concentration on Moses and the exodus events, De vita Mosis comes close in some respects to being a kind of Jewish parallel to a Christian Gospel.

What precisely the function of the miracles is in Philo’s portrayal of Moses will be discussed in the next section; the immediate issue is the nature of Philo’s portrayal of Moses, and to a lesser extent Aaron, as miracle-workers. One may, perhaps, question how far Philo in fact portrays Moses (or Aaron) as a miracle-worker at all. Although a large number of miracle-stories are narrated in De vita Mosis, it is God rather than Moses (or Aaron) who is said to work them. In Kahl’s terminology, Philo portrays Moses as a significant PNP and MNP, while making it clear that it is God who is the (ultimate and transcendent) BNP throughout.

Moses’ first encounter with Yahweh, the burning bush (Vit. Mos. 1.65–71), is clearly a ‘miraculous portent’ (71) brought about by God, not Moses. During the ensuing interview, God promises to give Moses three signs, the first two of which (rod-snake and blanched hand) Moses tries out at God’s bidding, the third (water into blood) he believes will happen on account of ‘the proofs he had already been shewn in the miracles of the hand and the staff, (82). When Aaron reproduces the first miracle in Pharaoh’s court, and his rod snake devours those of the magicians, even the most ill-disposed are compelled to recognize that the miracle was wrought ‘by some diviner power’ (94), and Philo adds that in these events ‘God had shown His will by the proofs of signs and wonders’ (95). On account of the continuing stubbornness of the Egyptians, however, it becomes necessary to punish them by ten miraculous chastisements. Philo makes it plain from the start that it is God who is behind their planning and execution:

God’s judgement was that the materials which had served to produce the world should serve also to destroy the land of the impious; and to show the mightiness of the sovereignty which He holds, what He shaped in His saving goodness to create the universe He turned into instruments for the perdition of the impious whenever He would. He distributed the punishments in this wise: three belonging to the denser elements, earth and water, which have gone to make our bodily qualities what they are, He committed to the brother of Moses; another set of three, belonging to air and fire, the two most productive of life, He gave to Moses alone; one, the seventh, He committed to both in common; and the other three which go to complete the ten He reserved to Himself (Vit. Mos. 1.96–97).

When Philo says that God assigned seven of the plagues to Moses and Aaron but reserved three to Himself, he does not mean that God acted as BNP only in the last three of them, but rather that Moses and Aaron were to act as MNP in the first seven, while God would act without a mediator in the remainder. The ensuing narrative ensures that the reader does not lose sight of this point. Aaron smites the river with his staff ‘at the command of God’ (99). The second plague (frogs) is initiated in like manner ‘at God’s command’ (103) and the frogs are subsequently removed when the Hebrews jointly act as PNP and make ‘intercession with God’ (105). The third plague is clearly instituted by God, for according to Philo, ‘Then God stayed from using water to afflict them, and used the earth instead; but appointed the same minister of chastisement, who once more, when bidden, struck the ground with his staff, when a stream of gnats poured forth’ (107). Moreover, Philo goes on to discuss why God should choose to punish the land through’ such petty and insignificant creatures’ (109), a discussion which incidentally diminishes Aaron’s role as MNP.

Admittedly, when Moses takes over from Aaron, we are told that ‘he [i.e. Moses] began to cause disturbance in the air’ (leading to the hailstorm; 114), as if Moses were the BNP here. But a little later on, once the destructive nature of this storm became apparent to the Egyptians, ‘they thought, as indeed was the case, that divine wrath (μηνιμάτων θείων) had brought about these novel happenings’ (119). When Moses next summons the plague of locusts, this is expressly ‘at God’s command’ (120). When the following plague, the darkness, is introduced it is not explicitly said who brings it about; the plague is just said to arise (ἐπιγίνεται, 123). It is only dispersed, however, when Moses prays to God (125); here again Moses acts as PNP and God as BNP. The seventh plague is instituted by Moses and Aaron together, but when they take ashes from the furnace and scatter them to produce the boils they are expressly described as acting ‘at God’s bidding’ (127). The final three plagues, Philo states, ‘were self wrought, without any human agent’ (130).

This does not essentially depart from the emphasis of the biblical account, in which Yahweh is clearly the BNP with Moses and Aaron acting as his human agents, so one cannot say that up to now Philo has deliberately subordinated Moses’ miracle-working activity to God; it would be more accurate to say that he has retained the biblical view.

God remains the BNP throughout Philo’s account of the exodus and wilderness events. Prior to the Red Sea crossing Moses predicts that God will find the trapped Israelites a way out of their plight (173–74), and when Moses subsequently strikes the sea with his staff it is once more at God’s command (177). Once the Hebrews are safely across but see their erstwhile pursuers drowned, they appropriately sing ‘hymns of thanksgiving to God’ (180). When the thirsty Israelites arrive at springs in the desert, only to find the water undrinkable, Moses supplicates God (184) and God shows Moses a tree and bids him cast it into the water to sweeten it (185). Although Moses subsequently becomes indignant when the people grumble for lack of food (196), God in his mercy and benevolence (198) devises ‘new and strange forms of benefaction’ (199) and sends the manna (200). Moses’ only explicit role in this instance is, under divine inspiration, to explain the origin of the heaven sent food (201). The ensuing account of the quails suggests that these were provided purely on God’s initiative (209). When Moses subsequently strikes the rock with his staff to bring forth water, he does this expressly ‘under inspiration (θεοφορηθείς)’ (Vit. Mos. 1.210).

Thus far Philo has been presenting Moses under the office of king. In Philo’s portrayal the miracles associated with him take place, not because he wishes to portray Moses as a thaumaturge, but because, given that God worked great signs and wonders at the time of the exodus, it was appropriate that the human intermediary should be the leader of his people. When Philo goes on to portray Moses as law giver, priest and prophet, there continue to be miracles, and God continues to be the BNP behind them. A miraculous fire spontaneously consumes the sacrifice offered by Moses and Aaron (Vit. Mos. 2.154). Aaron’s rod blossoms to confirm his appointment to the priesthood (179) after Moses has prayed for God’s help in the matter (177) and God instructed him to take a rod for each tribe (178). Moses, in his role as prophet, predicts what is about to take place at the Red Sea (251–52), but the miraculous parting of the waters then takes place ‘through the might of God (θείαις δυνάμεσι)’ (253), without any further activity on Moses’ part. It is also as prophet that Moses predicts the fate of Korah and his co rebels (281); the earth then indeed swallows up ‘the tents of the impious’ (282) while thunderbolts destroy a further ‘two hundred and fifty men who had led the sedition’ (283), but all this is the work of God, not Moses, for ‘The quick succession of these punishments and their magnitude in both cases clearly and widely established the fame of the prophet’s godliness, to the truth of whose pronouncements God Himself had testified’ (284).

When Philo refers to these miraculous deeds elsewhere, he says nothing to contradict the view that God is the BNP, whereas Moses’ role is merely to petition God’s aid, announce God’s actions, or act as the mediator of God’s power. For example, in describing the Red Sea miracle at Vit. Cont. 86 Philo writes, ‘For at the command of God the sea became a source of salvation to one party and of perdition to the other.’ Incidental references to the other Mosaic miracles also imply that it is God who worked them. According to Dec. 16, the reason for the law being given in the desert was that the miraculous provision of food and drink there should show the laws to be pronouncements of God (which implies the miraculous provision is seen as the act of God). De specialibus legibus 2.199 explains the fast after the ingathering as partially a reminder of the wilderness period when God provided food and water, whereas at the feast of baskets the celebrant is to recall how God ‘confounded their assailants with signs and wonders and portents and all the other marvels that were wrought at that time’ (Spec. Leg. 2.218). At Virt. 45–46 it is God who helps the Hebrews win an easy victory over the Midianites. At Det. Pot. Ins. 177 it is God who wrought the three signs given to Moses to convince his audience. As important as Moses is for Philo, when Philo looks back to the exodus events he regards the signs and wonders wrought at the time as acts of God, not deeds of Moses.

That said, there are two passages in De vita Mosis 1 where Philo appears to assign a more active role to Moses in miracle working. The first is at Vit. Mos. 1.155–56:

For, since God judged him [Moses] worthy to appear as a partner of his own possessions, He gave into his hands the whole world as a portion well fitted for His heir. Therefore, each element obeyed him as its master (τοιγαροῦν ὑπήκουεν ὡς δεσπότη τῶν στοιχείων ἕκαστον), changed its natural properties (ἀλλάττον ἥν εἶχε δύναμιν) and submitted to his command, and this perhaps is no wonder (καὶ θαυμαστὸν ἵσως οὑδέν). For if, as the proverb says, what belongs to friends is common, and the prophet is called the friend of God, it would follow that he shares also God’s possessions.

On the face of it this represents Moses as a BNP, since God shares everything with his friend, including the ability to make the ‘elements’ obey him.22 Holladay, however, argues that by στοιχείων Philo meant, not physical elements, but rather spiritual ones, that is that Moses had such control over his inner elements that he was able to attain perfect virtue, wisdom and citizenship of the whole world:

Immediately upon mentioning the catalogue of virtues [V Mos. 1.154], Philo says that God gave Moses τὸν πλοῦτον τῆς φύσεως, i.e. mastery over φύσις, which the Stoics conceived of materialistically. This applied not only to corporeal substances in the generally understood sense, e.g. trees and sealing wax, but even to ostensibly non-material entities, e.g. the soul or God. Given this metaphysics, virtue and vice were also conceived of as material substances, which meant that the practice of virtue or vice could be described in terms of Stoic physics, which is precisely what Philo does in this passage.

This suggestion avoids the difficulty that nowhere else (apart from the second passage we are about to look at) does Philo suggest that Moses was made a BNP. It also takes account of the context, which is concerned with Moses’ virtues as ruler, his partnership with God, and his citizenship of the world, but not with the working of miracles. Nevertheless, Holladay’s argument is vulnerable to two serious objections. The first is that Philo does not actually employ the phrase τὸν πλοῦτον τῆς φύσεως ‘immediately upon mentioning the catalogue of virtues’. The phrase ‘the wealth of nature’ appears in the LCL translation at 153 (and thus not ‘immediately upon’ anything said at 154), but it is not represented by the precise expression τὸν πλοῦτον τῆς φύσεως in the Greek; the translator has understood τὸν πλοῦτον to be implied from the previous sentence, so that ‘wealth of nature’ is here contrasted with material wealth. It is far from clear that the (implied) πλοῦτον means ‘mastery over’ (as opposed to possession) here, and in any case the phrase occurs a little too far from the disputed passage at 156 to fix its meaning in the way Holladay suggests. Neither, incidentally, does Philo actually say here that God gave Moses [τὸν πλοῦτον] τῆς φύσεως; Holladay has apparently conflated the mention of the wealth of nature at 153 with what Philo does go on to say at 155, namely that God gave Moses the wealth of the whole land and sea etc. (but not explicitly of ‘nature’).

The second objection is that although Philo is influenced by Stoicism to some extent, he by no means goes along with the Stoic metaphysics that Holladay describes. Philo is clear that God and the intelligible world are not material entities, and it seems unlikely that he would have envisaged either the higher soul or the virtues in this way either, since his Platonic depreciation of visible, material reality would naturally oppose him to such a view. Besides, Holladay cites no evidence that Philo took such a Stoic/materialistic view of virtue elsewhere in his writings.

The question is then whether Holladay’s basic suggestion can be salvaged without resorting to Stoic metaphysics, and this in turn depends on whether what Philo says at Vit. Mos. 1.156 can be translated in a way less inclined to suggest the power to work physical miracles than does the Loeb translation. A slightly more literal rendering of the Greek in the critical sentence of Vit. Mos. 1.156 might run, ‘Therefore each of the elements gave obedience as to a master, changed the power that it had and obeyed the commands.’ The range of possible meanings of δύναμις is broad enough to take in ‘power’ in some non-physical sense. It is the word στοιχεῖα that is more troublesome. It can mean ‘elements’ in the physical sense, or in connexion with grammar or logic, or fundamental principles of learning; it can also mean heavenly bodies or, conceivably, elemental spirits, but there is no obvious parallel to its meaning something like ‘elemental forces of the passions’, which is what would be required for a psychological reading here. Worse still, Philo has just used the word στοιχεῖον in what can only be a physical sense, where he says, ‘God rewarded him by giving him instead the greatest and most perfect wealth. That is the wealth of the whole earth and sea and rivers, and of all the other elements (στοιχεῖα) and the combinations which they form’ (155). For Philo then to use στοιχεῖον in some different and idiosyncratic sense two sentences later would be needlessly misleading, to say the least. The understanding of the disputed sentence as a reference to the performance of physical miracles must therefore stand, whatever the difficulties.

The argument therefore seems to have come up against an exegetical impasse. On the one hand there is a sentence which, for all its difficulties, can only mean that God granted Moses dominion over the elements of nature, apparently as a BNP, while on the other hand in no incident that Philo narrates does he portray Moses acting as such, but consistently subordinates Moses as the PNP or MNP for or through whom God acts. Before trying to resolve this impasse it will be helpful to move on to the second passage where Philo apparently makes Moses an autonomous thaumaturge.

When the Israelites became desperate for drinking water a second time, ‘Moses, taking that sacred staff with which he accomplished the signs in Egypt (διʼ ἧς τὰ κατʼ Αἴγυπτον ἀπετέλεσε σημεῖα), under inspiration smote the steep rock with it’ (210). Logically and grammatically, the subject of the relative clause can only be Moses. In other words Philo appears to be saying that it was Moses who performed the signs in Egypt by means of his rod. Yet the apparent difficulty here may provide the means of escaping from the previous impasse. In his actual description of the events in Egypt, Moses’ wielding of his staff is clearly subordinate to God’s directions. Philo cannot mean to contradict that here, he must mean that Moses performed signs with his staff as a mediator of numinous power the source of which was God. But if he can thus be expressing himself loosely here, he could equally well be expressing himself loosely in the first problem passage, so that what Philo must really mean by saying that God granted Moses authority over the elements was that he granted him authority only as MNP. The elements obey Moses’ command when Moses acts as God’s lieutenant, but when Moses’ acts as God’s lieutenant he is himself obeying his divine general’s commands. Despite appearances, therefore, these two problem passages do not contradict the contention that Philo portrays Moses as PNP and MNP.25

Philo also mentions a few miracles from before the Mosaic period. These begin right from the creation, since Philo notes that, ‘after a fashion contrary to the present order of Nature’ God first created trees ready-laden with food (Op. Mund. 40) and caused the earth to ‘put forth plants and [bear] herbs before the heaven was furnished’ (Op. Mund. 47). Given that Philo elsewhere (Vit. Mos. 1.212–23) justifies the miraculous on the basis of God’s power as Creator, these oddities may readily be regarded as miracles. Although Philo in one place dismisses the talking serpent in Eden as a mere myth (Agr. 96–97), in another place he apparently accepts it as literally true by explaining, ‘when some miraculous deed is prepared, God changes the inner nature’ (Quaest. in Gen. 1.32). Again, Philo stresses that the flood at the time of Noah was of an unprecedented nature (Abr. 42) and argues that the πνεῦμα that disposed of it again must have been a divine spirit rather than a wind since ‘that such (an amount of water) should be cleared out by the wind is not fitting, likely or right; but, as I said, (it must have been done) by the invisible power of God’ (Quaest. in Gen. 2.28). Philo also observes that the apparently instantaneous regeneration of the earth after the flood (Gen. 8:14–19) also required a miracle: ‘And do not wonder that the earth, given one day, grew all things through the power of God … for all things are possible to God’ (Quaest. in Gen. 2.47). Moreover, he repeatedly stresses the miraculous nature of Sarah’s conceiving Isaac in old age (Quaest. in Gen. 3.18, 56; 4.17; Abr. 111–12) just as he stresses the miraculous nature of the three angels who came to announce what would take place (Abr. 118). Again, the punishment of Sodom is clearly a miracle for Philo; even though he says, ‘I have given these details not in order to describe the unprecedented calamity of God’s mighty working’ (Abr. 142) he nevertheless regards the rain of fire as a startling and extraordinary punishment, newly created for the purpose of punishing the sin of Sodom (Abr. 137; cf. Quaest. in Gen. 4.51). Again, although Philo normally allegorizes the punishment of Lot’s wife (e.g. Leg. All. 3.213; Ebr. 164; Fug. 121), at Quaest. in Gen. 4.52 he treats it as a literal (and cautionary) miracle as well.

On the other hand, Philo minimizes Abraham’s role in the incident involving Pharaoh and Sarah. In Philo’s account, God acts purely on his own initiative to save Sarah from defilement (Abr. 96). To be sure, this is in accord with Gen. 12:17, but given what the Genesis Apocryphon makes of this story, it is significant that Philo makes no move towards turning Abraham into a PNP or MNP by having him pray to God.

Two general points emerge from this survey of Philo’s treatment of the miracles in Genesis. The first is that he is not shy of them; if they are narrated or even suggested in the biblical text he is willing to take them up and emphasize their miraculous nature, not only, perhaps, because they appear in the sacred text, but also to magnify the power of the Creator God for whom all things are possible. The second is that in these Genesis miracles it is always God who acts, not simply as BNP, but as BNP without any corresponding MNP or PNP. Given that no other human figure in Philo’s writings is given a role in the working of miracles, Moses’ (and Aaron’s) working of miracles, even as PNP or MNP, becomes all the more significant. No other figure in Philo’s writings acts as God’s partner in miracle-working even to this extent. This puts the problem passage at Vit. Mos. 1.156 into perspective; even if it only makes Moses a PNP/MNP, it still makes him a virtually unique figure (unique, that is, apart from his brother) in Philo’s writings. But this, of course, reflects the biblical estimate of Moses given at Deut. 34:10–12.

Nothing in Philo’s writings indicates that he knows of any miracles after the Pentateuchal period. This does not mean that he denied them; they are simply not mentioned. In particular, there are no accounts of miracles in the few writings that deal with Philo’s own time. Philo presents Flaccus as being duly punished by God in the eponymous book, but there is nothing to suggest that the punishment is at all miraculous, as there is nothing to suggest that Philo was going to narrate a miraculous end for Caligula in the Legatio. Neither does he describe any miraculous happenings among the Essenes or Therapeutae; but this is a point to which the next section will return.

The Function of the Miraculous in Philo

As with Josephus, Philo’s narratives include punishment, deliverance and provision miracles, as well as some minor categories. In place of the category of authenticating miracles found in Josephus’s writings, however, in Philo’s writings one should speak of evidential miracles.

The reason for this change of nomenclature is as follows. In Josephus one can identify a number of miracles, which he tended to call σημεῖα, the purpose of which was to authenticate a prophet. Philo does not adhere to a similar terminology (even though, following Septuagintal usage, he sometimes applies the word σημεῖον to the same events as Josephus). On the other hand, Philo frequently stresses that a particular miracle is a proof; but when a miracle functions as a proof, it is generally not to authenticate God’s human representative but to demonstrate God’s activity or will. We have already seen that miracles tend to have an evidential value for Philo; we may now run through some examples to indicate what Philo typically regards them as proving.

As in Exodus and in Josephus, when God commissions Moses at the burning bush he offers him three signs to convince the doubters. In the context of Philo’s account, however, Moses’ status is not at issue (even though God is responding to his fear of being thought a deceiver); God directs Moses to tell his audience that the one who sent him is ‘He Who is’, or, if they insist on some title, that he is the God of Abraham, Isaac and Jacob. It is directly after this that God declares, ‘And, if they still disbelieve, three signs which no man has ever before seen or heard of will be sufficient lesson to convert them’ (Vit. Mos. 1.75–76). Authenticating the sender may also serve to authenticate the one sent, but the emphasis here is more on demonstrating the reality of He Who is than on accrediting Moses as his agent.

Correspondingly, when Moses’ brother performed the first sign, the triumph of his snake over the others convinced the onlookers, not of Moses and Aaron’s status, but that these events were ‘brought about by some diviner power to which every feat is easy’ (94). Similarly, when the people grumbled against Moses on account of the lack of food in the wilderness, Moses, according to Philo, was not so much concerned for his own position as indignant that ‘after experiencing strange events outside the customary without number’ the people should still be failing to ‘put their trust in him of whose unfailing truthfulness they had received the clearest proofs’ (196). Philo’s Moses thus regards the previous miracles as evidence primarily for God’s trustworthiness. Again, the fire that consumes the sacrifice offered by Moses and Aaron serves to provide ‘the clearest proof that none of these rites was without divine care and supervision’ (Vit. Mos. 2.154), in other words, that the sacrifice was being offered in accordance with God’s will. Admittedly Moses’ own probity is at stake in the incidents that lead to the blossoming of Aaron’s rod (174–79), and in this instance Philo expressly states that Moses ‘besought God to shew them by clear demonstration that there had been no dishonesty in his choice of persons for the priesthood’ (177). On this occasion, then, the miracle does serve to accredit God’s agent. At 2.261, however, where Moses is again said to be indignant at the people’s failure to learn obedience from the miracles they have witnessed, the real issue is that they are not obeying God, even if it is Moses who has conveyed God’s commands about the collection of the manna (259). At 262, in connexion with the same incident, Philo remarks that ‘the Father confirmed the utterance of the prophet with two most convincing proofs’. This could be taken as an instance of signs authenticating Moses. But what God is said to confirm is not Moses’ status but his utterance (τὸ λόγιον), which is an expression of God’s own will. The confirming proof of this utterance is not some extrinsic sign, but rather the fact that what happens to the manna is in accordance with what Moses said. In the broader context Philo is discussing Moses’ work as a prophet, and his argument is not that God performed signs and wonders in order to authenticate his prophet, but that whatever Moses prophesied invariably came to pass (even if it involved a miracle). It is again in the context of prophecy that Philo treats the story of Korah’s rebellion. Here again, Moses’ own probity is clearly at stake, for Moses declares, ‘If the death they [the rebels] meet is in the ordinary course of nature, my oracles are a false invention; but, if it be of a new and different kind, my truthfulness will be attested’ (281). But the fate that Moses goes on to outline for Korah and his associates is presented as a prophetic prediction that comes true. Moses’ status as prophet is certainly at issue, but it is not so much being confirmed by a sign in Josephus’s sense as passing the test posed by Deut. 18:22.

The evidential use of miracle is also found outside De vita Mosis. In Philo’s view one reason for giving the law in the desert was that the miraculous provision of food and drink there might prove that the laws came from God (Dec. 16); the law-giving itself is likewise accompanied by miraculous phenomena (Dec. 33, 44). Philo’s comment on Gen. 16:1 at Quaest. in Gen. 3.18 suggests that the reason God waited until Sarah was barren with old age before making her conceive Isaac was to demonstrate that his providence was at work. At Quaest. in Gen. 4.51 the destruction of Sodom and Gomorrah is regarded, not only as a punishment for wickedness, but as proof of the power and sovereignty of God.

This miracle is nevertheless also a punishment. Yet Philo’s treatment of it in De Abrahamo shows a tendency in Philo, if not exactly to mitigate the punitive element, then at least to distance God from any taint of vindictiveness. For in Philo’s view, if the practices of Sodom had been allowed to go unchecked and to spread to the rest of humanity, the earth would have become depopulated (Abr. 136); God’s punishment of Sodom was thus motivated, not only by his wrath towards the sinner, but by his compassion on the rest of humankind (Abr. 137).

Similarly, when God punishes Pharaoh with ‘all manner of scarce curable plagues’ for abducting Abraham’s wife, Philo again prefers to stress his mercy and compassion rather than his judgment on sin: ‘And God, Who is kindly and merciful and shields the wronged, had pity for the strangers and plied the king with almost intolerable pains and grievous penalties’ (Abr. 96). After describing how the king’s indisposition discouraged him from any sexual misdemeanours, Philo again stresses that God acted to protect Sarah’s chastity, to evidence Abraham’s nobility and piety, and to safeguard their marriage, ‘from which was to issue not a family of a few sons and daughters, but a whole nation, and that the nation dearest of all to God’ (98). It is only when he goes on to consider the passage allegorically (99–106) that Philo discusses the king’s afflictions as a punishment, but by then the king has become an allegorical figure for ‘the mind which loves the body’ (103).

Philo can even stress God’s compassion in the plagues inflicted on Egypt. After describing the terrible suffering brought about when all Egypt’s water turned to blood, Philo states that ‘For seven days the terror reigned, until the Egyptians besought Moses and his brother, and they besought God, to take pity on the perishing. And He Whose nature is to show mercy changed the blood into water fit for drinking’ (Vit. Mos. 1.101). Again, one reason Philo gives for God’s punishing Egypt through so diminutive a creature as the gnat is that ‘God wished to admonish the inhabitants of the land rather than to destroy them’ (110). Finally, when Philo introduces the tenth plague, he explains, ‘This was the death of the Egyptians, not of the whole population, since God’s purpose was not to make a complete desert of the country, but only to teach them a lesson’ (134).

To be sure, this tendency to find forbearance and mercy even in the grimmest punishments is only a tendency. No reader of In Flaccum will suppose that Philo shares the modern squeamishness over the idea of a punishing God. Philo obviously believes that those punished by God deserve what they get (see, e.g., Vit. Mos. 1.236 where the eight discouraging and two faithful spies who report back on their reconnaissance of Canaan each get their just deserts; see also the description of the flood and the destruction of Sodom in Vit. Mos. 2.53–58 and Quaest. in Gen. 4.51; the destruction of Korah and his companions at Vit. Mos. 2.282–87; and the punishment of Lot’s wife at Quaest. in Gen. 4.52). Yet elsewhere in Philo’s writings there is a tendency to stress that God in his proper function acts as a benevolent creator, and to assign the work of punishment to angelic or human agents (e.g. Conf. Ling. 181). Something of the same tendency may have tempered his treatment of punishment miracles.

As with Josephus, and, indeed, with the biblical stories, a punishment miracle is often the shadow side of a miracle of deliverance. The drowning of the Egyptians in the Red Sea is an obvious case in point, since it secured the escape of the Israelites, just as the ten plagues inflicted on Egypt effected the Hebrews’ release from the land of bondage, or the single plague on an earlier Pharaoh effected Sarah’s release from his lustful clutches.

As in Josephus, the second kind of beneficent miracle in Philo is that of provision, which in Philo occurs solely as the provision of food and water to Israelites wandering in the wilderness (Vit. Mos. 1.185–86, 199–200, 206, 209, 210–12; 2.258–59; Spec. Leg. 2.199; 4.127–30; Dec. 16; Congr. 173). The obvious purpose of such miracles is to supply the need of the Israelites, though, as we have seen, Philo can also regard these miracles as having evidential value (Dec. 16); he can also describe them as the expression of God’s compassion (Vit. Mos. 1.199) or generosity (Vit. Mos. 1.209).

In Philo the minor categories of miracle include those of creation and recreation: trees created ready-laden with fruit (Op. Mund. 40), plants created before the heavens were furnished (Op. Mund. 47), disposal of the flood waters (Quaest. in Gen. 2.28), and renewal of earthly flora after the flood (Quaest. in Gen. 2.47). There are also a few miscellaneous miracles: the miraculous voice God uses to utter the ten commandments (Vit. Mos. 2.213 and Dec. 33), the talking serpent in Eden (Quaest. in Gen. 1.32), and the translation of Enoch (Quaest. in Gen. 1.86). The link between miracle and creation is further emphasized at Vit. Mos. 2.267 where the manna miracle is likened to the creation of the world (God calls up plenty in the desert just as he called up the world out of nothing).

One category of miracle that is missing from Philo is that of healing. Healing miracles are scarce enough in Josephus, but Philo has none at all. Admittedly there are not many healing miracles in the Old Testament, and even fewer in the Pentateuch, but Philo makes no use even of such as are implied in the text. He makes no miraculous use of the recovery from illness of the king punished for lusting after Abraham’s wife. Neither does he narrate Miriam’s recovery from (penally inflicted) leprosy (Num. 12:9–15) except to allegorize it (Leg. All. 1.76). He mentions the bronze serpent that Moses put on a pole to save people from snake-bite (Num. 21:9), only to dismiss the literal sense of the story as patently absurd in order to justify his allegorical interpretation in which the serpent represents self-control (Agr. 95–98).

Philo’s failure to mention miraculous healing is all the more striking given his readiness to talk about other types of healing, both physical (e.g. Leg. Gai. 106; Spec. Leg. 1.77; Leg. All. 3.178; Sacr. 70, 123; Deus Imm. 65; Conf. Ling. 22; Congr. 53; Somn. 1.112; Aet. Mund. 63) and spiritual (e.g. Vit. Cont. 2; Op. Mund. 155; Spec. Leg. 1.197, 237; 2.17; Virt. 26; Leg. All. 1.45, 70, 76; 2.79, 87; 3.36, 106, 124, 129, 215; Sacr. 70, 121, 123, 127; Det. Pot. Ins. 44, 110, 146; Deus Imm. 63, 67, 135; Mig. 124, 219; Rer. Div. Her. 299; Congr. 18; Somn. 1.112; 2.197; Quaest. in Gen. 2.29; 3.10; Prov. 2.18). If the predominance of the latter category suggests that Philo valued spiritual above physical healing, this is also suggested by some of his incidental remarks, for example, ‘Cowardice, too, is a disease graver than any that afflicts the body … unless it is healed by God’ (Virt. 26); those who face both ways flee for help to physicians, medicines and other created things rather than to God, ‘the one and only physician of soul-sickness’ (Sacr. 70); or perhaps most explicitly of all, in connexion with the Therapeutae:

they profess an art of healing better than that current in the cities which cures only the bodies, while theirs treats also souls oppressed with grievous and well-nigh incurable diseases, inflicted by pleasures and desires and griefs and fears, by acts of covetousness, folly and injustice and the countless host of the other passions and vices (Vit. Cont. 2).

Despite the tone of such remarks, Philo does not intend to depreciate either physical health or the medical profession. Indeed, he often employs medical analogies; and physicians, in addition to providing an analogy for healers of the soul, also appear as stock examples of skilled professionals who know their trade and perform a useful service (e.g. Sacr. 123; Deus Imm. 65), just as medicines and herbs are acknowledged as beneficial (e.g. Congr. 53; Aet. Mund. 63). Although Philo does not deny God’s role in healing, he expects the physician to be the channel through which God effects it, as he states explicitly:

God bestows health in the simplest sense, preceded by no illness in our bodies, by Himself only, but health that comes by way of escape from illness He bestows both through medical science and through the physician’s skill, letting both knowledge and practitioner enjoy the credit of healing, though it is He Himself that heals alike by this means and without them (Leg. All. 3.178).

If the three words ‘and without them’ appear to leave the door open to miraculous healing, Philo never exploits this opening. Given his other comments it is more likely that he regards spiritual health as conducive to bodily health. This may be the sense of the quotation from Vit. Cont. above, which suggests that the Therapeutae heal not only the body but also the soul. Some such thought may also lie behind the statement at Spec. Leg. 1.77 that donors bring first-fruits to the temple gladly, expecting that this payment will give them release from slavery or healing from illness; a virtuous act may help cause its own reward. It may also be indicated by the statement at Virt. 13 that ‘diseases of the body, if the soul is healthy, do very little harm’ Praem. Poen. 119 lists ‘complete freedom from disease’ and ‘efficiency of the senses’ as among the rewards promised to ‘those who take pains to cultivate virtue and set the holy laws before them to guide them in all they do or say’.

Philo incidentally implies that normal, medical means of healing prevailed among the Essenes. At Hyp. 11.13 he notes in passing that a sick Essene is nursed and tended by all at common expense, whereas Omn. Prob. Lib. 87 states that among Essenes the sick have the cost of their treatment met out of the common stock. This hardly suggests that any charismatic healing gifts were employed.

If Philo knows nothing of miraculous healing, he also says nothing about exorcism, not least because he acknowledges no evil spirits. On the odd occasion where he uses the word δαίμων he generally explains that it is the pagan equivalent of what Moses called ‘angels’ (Gig. 6; Somn. 1.141; Quaest. in Gen. 4.188) and that ‘souls and demons and angels are but different names for the same one underlying object’ with the result that one can be free from ‘that most grievous burden, the fear of demons or superstition’ (Gig. 16). Sometimes Philo wants to restrict ‘angel’ to the sense of a purely spiritual being whose abode is the air and whose function is to serve God (Quaest. in Exod. 2.13). But since ‘angels’ means ‘souls’, the word can include human souls. Those that remain bodiless, the servants of God, are thereby immune from wickedness, because they have not been imprisoned in the body (Conf. Ling. 177). But some bodiless souls created by God enter mortal bodies (Plant. 14). When Philo refers on one occasion to ‘evil angels’, at Gig. 17, the reference is almost certainly to wicked human souls. Philo’s polemic against superstition suggests that he is acquainted with popular demonology; he may well know of exorcists (though his writings provide no evidence for that), but he explicitly rejects the former and hence implicitly rejects the latter.

Although it is not my main concern, I should say a few words about the allegorization of miracle in Philo. This is not entirely confined to his allegorical writings. For example at Vit. Mos. 1.61–70 the fact that the bush was not consumed is taken as symbolizing that the suffering Hebrews would not be destroyed by their aggressors. Here there is a greater congruence between the symbol and the thing symbolized than is usual in Philo; and given that God is about to lend miraculous aid, it is appropriate that the visual symbol should itself be a miraculous occurrence.

This, however, is atypical of Philo’s allegories. In his esoteric writings, there is often little congruence between the surface meaning of the text and the allegorical meaning Philo finds in it. In particular, when he is allegorizing a miracle, the miraculous nature of the surface event is generally irrelevant to the allegorical meaning. For example, the Red Sea events, particularly the drowning of the Egyptians and the song of triumph ‘horse and rider he has cast into the sea’, are frequently allegorized as the overcoming of the wayward passions symbolized by ‘horse and rider’ (Leg. All. 2.102; Agr. 82; Ebr. 111; Sobr. 13; Somn. 2.269; cf. Ebr. 79). Moses’ serpent turning back into a rod is likewise taken to denote pleasure brought under control by discipline (Leg. All. 2.88), as its swallowing of the magicians’ rods shows that ‘arguments of sophists are devoured … by Nature’s many-sided skill’ (Migr. Abr. 85). Manna is allegorized as spiritual nourishment, wisdom, logos, or God’s words (Leg. All. 3.162; Fug. 137–38). The miracle of the waters of Marah represents chastisement as educative, or law given in a place of toil, or the toil that is the love of labour, the yearning or love of the good (Congr. 163–66; cf. Poster. C. 156); the tree that effected the sweetening is a figure for sweetening what is bitter in the soul (Migr. Abr. 36). When Lot’s wife becomes stone, she stands for a soul that God did not permit to repent (Leg. All. 3.213); she turns to stone because she lags behind and gazes round at old familiar things and remains among them like a lifeless object (Ebr. 164; cf. Fug. 121–22, Somn. 1.247–48).

One might suggest that in some of Philo’s allegorizations an outward, physical miracle is transformed into an inward, spiritual one; so that, for example, God’s miraculous aid can overcome the sickness of the soul (e.g. Spec. Leg. 1.282; Virt. 26). But we have already seen that Philo has a sharply defined concept of miracle for which he tends to use a small set of distinctive terms; the proposed ‘spiritual miracle’ does not conform to this concept; neither does Philo use his characteristic miracle vocabulary when he interprets miracles in this way. Moreoever, by interpreting miracles as figures for what might typically happen to the appropriate type of soul at any time, Philo strips them of their character as concrete, historical events. Within the allegorized interpretation, therefore, they cease to be ‘miracles’ in any sense relevant to the current investigation.

In addition to the foregoing discussion of what function individual types of miracle play in Philo’s writings, one may also enquire into the role of the miraculous in general in his work. For Josephus miracles were striking instances of the workings of God’s providence, understood in terms of retributive justice. Philo only rarely makes an explicit link between miracle and providence (e.g. at Quaest. in Gen. 3.18 and Vit. Mos. 2.261; in the latter the Greek expression is not πρόνοια, but ἐπιφρονοσύναις θείαις), but there is nothing to suggest that he would have denied such a link. Philo’s concept of providence, however, is not so much one of retributive justice worked out in history as God’s active care for his creation. Miracle is important to Philo as defining God as the biblical God who acts, rather than the Aristotelian Unmoved Mover or the Stoic world-soul. One reason Philo narrates miracle stories is because he finds them in the biblical accounts he is expounding, but by adopting them he also takes over the concept of God they imply. The miracles may also have a propagandist function, indirectly glorifying Judaism by showing what great things God did for its most eminent representative and for the people as a whole in the foundational events of the nation’s history.

Finally, one may ask how seriously Philo takes the miracles he narrates as actual, historical events. Philo’s inclusion of miracle stories in a work which purports to be an accurate life of Moses (Vit. Mos. 1.1, 4) must be taken as prima facie evidence for his believing them to be true. There is little reason to suppose that Philo doubted what he read in the sacred text, except where he explicitly rejects its literal sense in order to lead to an allegorical interpretation. This he seldom does in the case of miracles. Moreover, his defence of the possibility of miracle (e.g. Vit. Mos. 1.212–13), together with his repeated insistence that all things are possible for God, also indicates that he believed in their reality.

Philo occasionally attempts to offer a natural explanation of miracle. For example, when God shows Moses the tree to throw into the waters of Marah, Philo suggests that this tree had been ‘possibly formed by nature to exercise a virtue which had hitherto remained unknown, or possibly created on this occasion for the service it was destined to perform’ (Vit. Mos. 1.185). Similarly, when Moses strikes the rock with his staff to obtain water, Philo goes on to say, ‘It may be that the rock contained originally a spring and now had its artery clean severed, or perhaps that then for the first time a body of water collected in it through hidden channels was forced out by the impact’ (Vit. Mos. 1.211). The second of these certainly looks like a plausible explanation, but Philo at once undermines the idea that the event was merely natural by giving a spirited defence of miracles (212). Again, when the first explanation (at Vit. Mos. 1.185) is examined more carefully, it is not so non-miraculous. Given the link Philo elsewhere suggests between miracles and creation, a tree specially created for the occasion would have to count as a miracle. Even if one takes the first half of this explanation seriously, namely that the tree was ‘formed by nature to exercise a virtue which had hitherto remained unknown’, one would have to note, first that it must have been something of a miracle that Moses recognized its potential purpose (the text says that God opened his eyes to the tree and directed him in its use); secondly, that since Philo is unable to suggest what kind of tree it might be, its properties must have remained unknown since Moses’ time, so that it must have been a very exceptional kind of tree, if not unique; and thirdly, that it is at least possible that here as several places elsewhere when Philo says ‘nature’ he really means ‘God’. In any case Philo does not normally offer rationalizing explanations for miracles but instead tends to emphasize how miraculous they are. One or two half-hearted exceptions (even supposing they are genuine exceptions) do not suffice to overturn this general rule. Rather one must suppose that Philo, like Josephus, occasionally offered a naturalistic explanation whenever one happened to occur to him. In the main, however, Philo is as strong a believer in the reality of the miracles he relates as Josephus.


This chapter breaks more new ground than the last, since very little has been written specifically on miracle in Philo, apart from the article by Delling (in German). Although I have not disagreed with Delling to any great extent, I have addressed a slightly different set of questions. In particular, I have conducted my analysis of miracle in Philo in such a way as to aid direct comparison with the same theme in Josephus. This has shown that Philo’s concept of miracle is more sharply defined than that of Josephus, but that Philo shares both Josephus’s emphatic belief in the miraculous and his lack of interest in miracles of healing. The latter may partly be explained by Philo’s focus on the Pentateuch, but may nonetheless be significant given Philo’s readiness to discuss (non miraculous) healing in various contexts. Also of interest in Philo is the restriction to Moses, and to a lesser extent Aaron, as the only human figures through whom miracles are worked; elsewhere in Philo miracles are attributed directly to God.

Despite potential points of contact with the Gospels, it is the differences between Philo and the Evangelists that stand out the most. Whereas Philo’s Moses is given signs to perform to furnish objective proof of God’s backing, the Synoptic Jesus declines to perform any such signs. Whereas Mark’s Jesus spends himself in healing the sick and casting out demons, Philo admits no such miraculous cures. Above all, for Philo the great age of miracles is associated with the distant Mosaic past; that God might act decisively in the present through another miracle-working figure seems remote from Philo’s concerns. In common with Josephus, Philo tends to narrate types of miracle that are absent from the Gospels: punishment miracles, evidential miracles, and miracles of deliverance from human foes. Only in miracles of provision does he, like Josephus, overlap with a type of miracle found in the Gospels (notably the feeding stories: Mk 6:34–44; 8:1–9 and parallels, and perhaps also the Wedding at Cana, Jn 2:1–11, and the Miraculous Catch of Fish, Jn 21:1–6; Lk. 5:1–9).

Another text that may have been produced in Alexandria at around the time of Philo is the Wisdom of Solomon. In any case it is a text that bears sufficient similarities to Philo’s writings for it to be worth comparing their views on miracle. The next chapter will accordingly explore miracle and Wisdom.

Chapter 4

Miracle and Wisdom


The two great examples of post-biblical Second Temple wisdom literature are the Wisdom of Solomon and the Wisdom of Ben Sira. Although the first-mentioned of these texts was most probably the second to be written, it will be discussed first in this chapter, since, like Josephus and Philo, it exemplifies Greek-speaking Hellenistic Judaism, and may have come from roughly the same time and place as Philo (see below). Ben Sira, on the other hand, is something of a bridge text. In its Greek translation Sirach partially exemplifies Greek-speaking Hellenistic Judaism simply through having been translated into Greek in Egypt and absorbing some ideas from that milieu in the process. In its Hebrew original, on the other hand, the Wisdom of Ben Sira may well be taken as a fairly representative example of moderately conservative Hebrew-speaking Palestinian Judaism from the early second century bce. In its two forms the work thus provides a convenient transition from a Greek speaking to a Hebrew-speaking milieu in this survey of Second Temple literature.

The Wisdom of Solomon will be treated in much the same manner as Josephus and Philo (thereby facilitating comparisons), albeit on a smaller scale. For Ben Sira, however, the strict pattern of examining first the language and concept, then the distribution, and finally the functions of the miraculous will be abandoned as being too cumbersome for the nature of the text.

The Wisdom of Solomon

a. Introduction

The Wisdom of Solomon (hereafter Wisdom) is mainly concerned with matters other than miracle, but has a substantial section, in this case on the exodus, in which many biblical miracles are recalled. Moreover, Wisdom has some affinities with the writings of Philo, which make it worthwhile to compare the treatment of miracle in each.

It is now generally agreed that Wisdom was originally composed in Greek, the major (but far from only) manuscript witnesses being Codices Sinaiticus, Vaticanus and Alexandrinus. The date and provenance are less clear. Since Wisdom seems to be familiar with the Septuagint, the terminus a quo must be around 200 bce. David Winston has identified a number of words used in Wisdom not attested elsewhere until the early first century ce. This probably places Wisdom somewhere in the early Roman period, though it is impossible to be more precise.4 The text seems particularly occupied with the Egyptian worship of animals and with the punishment of the Egyptians at the exodus, and this has led many scholars to identify the place of writing as Egypt, or more specifically, Alexandria. Against that it could be argued that most Jewish writers, whatever their location, would look back to the exodus as the paradigmatic saving event and that it is the focus on the exodus that produces the preoccupation with Egypt in Wisdom, not necessarily an Egyptian provenance. Nonetheless, Wisdom would certainly fit an Alexandrian provenance quite well, not least on account of its affinities with Philo, and also because its rhetoric could well be directed to Alexandrian Jews chaffing at their loss of status with the onset of Roman rule.7 That Wisdom therefore came from roughly the same time and place as Philo would seem the best guess, even though it cannot be established beyond reasonable doubt.

Wisdom falls into three sections: chapters 1–5 form an address to the rulers of this world warning that righteousness and injustice will receive their due rewards in the next; chapters 7–9 form a praise of wisdom; and chapters 11–19 contrast how the same elements punished the Egyptians and saved the Israelites at the times of the exodus, via lengthy digressions on punishment and idolatry (12–15); chapters 6 and 10 form bridge passages linking one section to the next. The apparent differences between the sections led some earlier scholars to see the work as composite, but now it is generally (and rightly) regarded as a unity. The genre of the piece is still something of a puzzle, however. Some scholars see it either as aprotreptic9 or as an encomium. How closely it really fits either of these supposed Graeco Roman models is open to question, although Wisdom may have borrowed from both.11 But however it should be formally classified, its purpose seems to be mainly rhetorical. Wisdom may plausibly be understood as an attempt to persuade and equip educated Alexandrian Jews to remain loyal to their faith in the context of Hellenistic culture. That the function of the piece is mainly rhetorical rather than discursive may well account for its apparent lack of philosophical sophistication compared, say, with Philo, although Wisdom shows more signs of Greek philosophical influence than does Ben Sira. Moreover, what has been seen as an eclectic hotchpotch of Stoic and Platonic ideas may more sympathetically be seen as a reflection of Middle Platonism.

Whether Philo knew Wisdom or vice versa cannot be demonstrated, but, as has already been intimated, there are points of contact between Philo and Wisdom that suggest that some of Philo’s ideas were not completely peculiar to himself; it may be that both writers were drawing on a common stock of traditions. The following discussion makes frequent comparisons between Wisdom and Philo; to aid that, the treatment of miracle in Wisdom will be structured under the same three main headings as were used for Philo (and Josephus), even though in itself Wisdom might hardly merit quite such detailed treatment.

b. The Language and Concept of Miracle in the Wisdom of Solomon

Unlike Philo, Wisdom lacks any specialized miracle vocabulary; there are no words that regularly appear in the context of miracle, neither do the words that Wisdom does employ provide much clue concerning Wisdom’s concept of miracle.

The words Wisdom uses in connexion with miracle include σημεῖον, τέρας, θαυμαστός and παράδοξος, though none of these occurs frequently. Σημεῖον is used at Wis. 8:8 and 10:16 in the stock phrase ‘signs and wonders’ or ‘wonders and signs’. It also appears once in a non-miraculous context at 5:13. These instances also account for half the occurrences of τέρας. As for the other half, at 17:14 (lxx) τέρασιν … φαντασμάτων refers to the ‘monstrous spectres’ (17:15 rsv) that beset the Egyptians during the plague of darkness, while at 19:8 the Israelites who passed safely through the Red Sea had gazed on ‘marvellous wonders’ (θαυμαστὰ τέρατα). This in turn accounts for one of the two the uses of θαυμαστός, the other one being at 10:17 where Wisdom is said to have guided the Israelites ‘along a marvellous way’ (ἐν ὁδῷ θαυμαστῇ). A similar sentiment is expressed in different words at 19:5 where the Egyptians’ foolish decision to pursue the Israelites leads to the latter’s experiencing an ‘incredible journey’, παράδοξον ὁδοι πορίαν (lxx). Πσράδοξος is also used (in the superlative) in a miraculous context at 16:17, when it refers to the (most incredible) fact that the fire (i.e. lightning) that accompanied the plague of hail (and rain) became even more effective in the water (cf. Vit. Mos. 1.118). The third occurrence of παράδοξος in Wisdom, at 5:2, refers to the unexpected post-mortem salvation of the righteous as seen by the wicked. Although we shall return to them in connexion with Wis. 8:8, the terms σημεῖον and τέρας thus seem to have been borrowed from the Septuagint, while θαυμαστός and παράδοξος strike a very occasional note of wonder or surprise.

There are a few other places where Wisdom calls attention to the odd nature of a miracle. For example the weather conditions in the plague of hail are described as ‘strange’ (ξένοις, 16:16; cf. Vit. Mos. 1.114–17, where Philo explains why such weather would be unprecedented in Egypt). But it cannot be said that Wisdom lays much stress on the amazing or seemingly impossible nature of the miraculous. Perhaps this is simply taken for granted. It may also be implied by the emphasis on the omnipotence of God. For example Wis. 11:17–20 refers to God’s ‘all powerful hand, which created the world out of formless matter’ (11:17) and goes on to suggest that God could have sent or specially created creatures far more terrible than those he actually employed to punish the Egyptians. Verses 11:21–22 similarly emphasize God’s power: ‘For it is always in thy power to show great strength, and who can withstand the might of thy arm? Because the whole world before thee is like a speck that tips the scales; and like a drop of morning dew that falls upon the ground’. The relatively mild polemic against those who search creation but stop to worship it rather than its (so much greater) creator (13:1–9) also points in this direction, as do the powerful actions attributed to wisdom (10:1–11:1), and statements addressed to God such as ‘For thou hast power over life and death’ (16:13) and ‘To escape from thy hand is impossible’ (16:15).

Yet many of the passages that emphasize God’s power at once go on to suggest a self limitation in the exercise of this power. For example, Wis. 11:20 recalls, ‘But thou hast arranged all things by measure and number and weight’, and at 11:23 it states, ‘But thou art merciful to all, for thou canst do all things, and thou dost overlook men’s sins, that they may repent.’ The first of these statements could perhaps be taken to mean that God has created an orderly cosmos, and so respects the order he has created that he refrains from purely arbitrary interventions. But ‘measure and number and weight’ may more specifically be a reference to proportionality in punishment, the point being that God fits the punishment to the crime and so refrains from inflicting some harsher punishment (in this case by creating some terrible monster to send against the Egyptians).17 The second statement is concerned not so much with order and justice as with mercy. God can do all things; the implication is that God could inflict any punishment that occurred to him, but that forbearance is also among the things of which he is capable, and it is this that he in fact chooses.

The most notable attempt to make miracles acceptable within a framework of popular philosophy occurs at 19:18:

For the elements changed places with one another,

as on a harp the notes vary the nature of the rhythm,

while each note remains the same.

The ‘elements’ in this case are fire, earth and water (air is not explicitly referred to). Verse 19:19 refers to the interchange of land and water creatures, each moving in the other’s environment. The reference here is presumably to the Israelites’ passing through the sea as if on dry land, and perhaps to the plague of frogs that infested the land of the Egyptians. Verse 19:20 similarly discusses fire and water: the rain failed to quench the lightning (fire) in the plague of hail and the fire failed to destroy the manna that could be easily melted by the sun’s rays (perhaps water and fire are in turn each supposed to behave like air, allowing the fire to burn but not melting the manna, so that there is an implicit reference to the fourth element here). The imagery of the harp is a little obscure; Winston suggests that the reference is to changing key (or rather, mode) rather than rhythm, but the underlying point is clear enough.19 Just as one can change the nature of a melody by varying the rhythm (or mode) while employing the same notes, so the properties of elements may be interchanged or transposed, the link between the two ideas being the Stoic theory of the tension and relaxation of the elements. This Hellenistic physical theory is thus employed in an (albeit forced) attempt to render the miraculous credible.21

Wisdom’s theory of miracles is similar to Philo’s (see Chapter 3, Section 2 above), but the emphasis is different. Although, as we have seen, Wisdom does allude to the omnipotence of God, it does not use it to justify the occurrence of miracles in the way Philo does (partly, no doubt, because it is a different type of writing). Wisdom is also less concerned than Philo to demarcate miracles as special events; their strangeness is sometimes referred to, but it is not stressed, and there is no clear-cut miracle vocabulary that would indicate a conscious awareness of a distinct class of events. Wisdom would no doubt agree with Philo that miracles are acts of God (this will be argued more fully in the next section) that objectively occur (this issue will be discussed further in the section after next). On the other hand, the tendency of Wisdom’s treatment of miracles in the exodus section is less to stress their historicity than to fit them into the author’s scheme of rewards and punishments (a point to which we shall again return). But if this makes the miracles seem more abstract, it does not do so in the manner of Philo’s allegorization, rather it turns them into ‘ideal types’ of God’s action in history, and so leaves it more open than Philo is willing to state that God may act thus again.

c. The Distribution of Miracles in the Wisdom of Solomon

The miracles to which Wisdom alludes are taken almost exclusively from the exodus story (including the plagues, the Red Sea crossing, and the wilderness wanderings), and are mainly described in chapters 11–19. Chapter 10 recalls the actions of wisdom in the history of Israel up to the Exodus, and so refers to such miracles as the destruction of Sodom, Gomorrah and the other three cities of the plain (10:6) and to Lot’s wife being turned into a pillar of salt (10:7), but there is no emphasis on these events as miracles. Wis. 5:17–23 describes mighty acts that the Lord will do on behalf of his people, but in terms that recall the exodus miracles. Indeed, this passage anticipates the later and longer exodus section both by stressing that God employs his creation to smite his enemies (5:17, 20) and in the type of divine warfare envisaged: lightning, hailstones, sea, rivers and wind (5:21–23), all of which could be related to the Red Sea crossing or the plagues. To be sure, the descriptions are vague enough to be taken as allusions to other biblical incidents; for example, although the statement, ‘hailstones full of wrath will be hurled as from a catapult’ (5:22) could certainly refer to the plague of hail, it could also refer to the hailstones that aided Joshua against the Amorites (Josh. 10:11). But overall, it appears to be the exodus miracles Wisdom has mainly in mind. The passage thus serves to link this section of the book with the subsequent section on the exodus, but also suggests that the section on the exodus can be read paradigmatically as exemplifying the type of action God is always prepared to take. The closing verse (19:22) is thus not only a concluding doxology, but a promise that what God has done before he may well do again: ‘For in everything, O Lord, thou hast exalted and glorified thy people; and thou hast not neglected to help them in all times and in all places.’

The agent of the miraculous in Wisdom is nearly always either God or wisdom. Since wisdom functions as a form of divine agency in Wisdom, this probably comes to much the same thing (and in any case, as we shall shortly see, the same miracle can be attributed to both). At 5:17–23 it is God who ‘will take his zeal as his whole armour, and will arm all creation to repel his enemies’. Likewise from 11:4 onwards in is God, addressed in the second person, who performs virtually all the miracles described (exceptions will be noted below). In particular he provides the Israelites with water from a rock, in contrast to turning water into blood for the Egyptians (11:4–8); he sends animals to punish the Egyptians but quails to feed the Israelites (16:1–4), locusts contrasted with serpents (11:5–14), rail, hail and fire contrasted with manna (16:15–23), and a pillar of fire (18:3) in contrast to the plague of darkness (17:1–21); and he is responsible for the death of the Egyptian first-born (18:5–14) and the Israelites’ safe passage through the Red Sea (19:6–7).

At 10:18–19, however, it is wisdom that is said to have brought the Israelites through the Red Sea while drowning their enemies. The fact that the same miracle is attributed both to God and to wisdom is a good indication that for Wisdom there is little difference between attributing agency to one or the other. The theoretical distinction may be that wisdom is the means by which the transcendent God acts in the world. Elsewhere it is less clear precisely what wisdom’s agency amounts to in miraculous events. At 10:6 she is said to have rescued a righteous man from the fire that destroyed the five cities, but it is not said who sent the fire. At 11:1 she is said to have prospered the Israelites’ works ‘by the hand of a holy prophet’. If this refers not simply to the sagacity Moses exercised in leadership but to the miracles wrought through him then wisdom is here a BNP and Moses the MNP.

This relation between wisdom and Moses is in any case stated explicitly at 10:16. ‘She entered the soul of a servant of the Lord, and withstood dread kings with wonders and signs’. ‘Dread kings’ seems to be a way of making a general reference to Pharaoh in a manner sufficiently vague to suggest that wisdom may act similarly through other servants of God against other bad rulers. The phrase ‘wonders and signs’ clearly reflects the stereotyped language often used to refer especially to the exodus miracles. The statement that wisdom entered Moses’ soul is particularly noteworthy. It could mean simply that Moses’ natural wisdom was enhanced by divine wisdom, but the working of miracles is not normally a function attributed to wisdom in other wisdom texts; miracle-working is generally thought of not as a skill but as a power. If Wis. 10:16 means that wisdom empowered Moses to work miracles, wisdom is thereby conceived along the lines of the spirit that gives charismatic inspiration to prophets (cf. Isa. 63:11–14 where many of the acts attributed to wisdom in Wisdom are attributed to God’s Holy Spirit). This is not impossible: Wis. 1:6 states ‘Wisdom is a kindly spirit’, and 7:25 that ‘she is the breath (ἀτμὶς) of the power of God’. Since the spirit could also be seen as a means or mode of God’s operation in the world, an equation of wisdom and spirit could well be intended here. What is then odd about 10:16 is that elsewhere in Wisdom the equation of wisdom and spirit seems to nudge the latter concept away from its more charismatic associations towards a λόγος concept, that is, towards a rational ordering principle (e.g. Wis. 1:7). This is, however, in keeping with one indication of a reverse tendency to assimilate the λόγος concept to Jewish views on the role of angelic powers. At 18:15–16 the role of the angel of death, slaying the Egyptian firstborn, is given to God’s ‘all-powerful word’ (ὁ παντοδύναμός σου λόγος). This slightly curious mutual assimilation of word and spirit (and angel) is presumably a by-product of Wisdom’s attempt to marry Jewish and Greek traditions.

The other place where wisdom is connected with miracle, and in a manner that is not immediately clear, is at 8:8, where ‘she has foreknowledge of signs and wonders and of the outcome of seasons and times’. According to James Reese, ‘The common biblical phrase ‘signs and wonders’ here does not refer to God’s mighty actions but to the natural laws governing the universe.’ Unfortunately, Reese gives no reason for this statement.28 Perhaps he takes ‘seasons and times’ (καιρῶν καὶ χρόνων) to refer to the ordered march of natural events, but it is surely just as likely to refer to the occurrence of significant events in history. Moreover, it seems strange to take ‘signs and wonders’ to refer to ‘the natural laws governing the universe’ at 8:8, when the only other occurrence of a similar phrase, ‘wonders and signs’ at 10:16, clearly refers to the exodus miracles, and does so using a biblical cliché (e.g. Deut. 29:2 lxx). neb translates, ‘she can read signs and portents, and can foretell the outcome of events and periods’. This makes one verb in Greek, προγινώσκει, do duty for two verbs, ‘read’ and ‘foretell’ in the proposed English translation, where only the latter is a fair rendering of the Greek verb in question. The neb translation can nevertheless be defended on the grounds that it fits the context well. The previous two lines run, ‘she knows the things of old, and infers the things to come; she understands turns of speech and the solutions of riddles’ (rsv). If the structure of 10:16 is taken as one of chiastic parallelism, then knowing the things of old and inferring the things to come would correspond to foretelling the outcomes of times and seasons, whereas deciphering signs and portents would correspond to deciphering turns of speech and riddles. The Greek σῆμεια καὶ τέρατα προγινώσκει must then be taken as an elliptical way of saying, ‘she foreknows what signs and portents portend’, the sense of which is given well by neb. The alternative is to take ‘signs and wonders’ simply as a stereotyped phrase meaning ‘miracles’ and to take the last two lines as together meaning that wisdom has foreknowledge of the divine action in history: miracles (i.e. God’s mighty acts) and the outcome of times and seasons. This alternative has the advantage of being closer to the plain sense of the Greek than neb, but loses the close parallelism between deciphering portents and deciphering riddles. Perhaps the turns of speech and riddles would then have to be understood as those uttered in prophecy. But whichever of these two ways the text is construed, wisdom is not represented in this verse as an agent of the miraculous, but simply as its predictor or interpreter.

We have already noted one or two places where Moses is represented as an MNP (10:16 and 11:1). There is one other place where a man is given a miracle-working role, namely where Aaron (‘a blameless man’) puts an end to a plague in the desert: ‘he brought forward the shield of his ministry, prayer and propitiation by incense’ (18:21). In other words, Aaron’s role is quite explicitly limited to that of PNP.

Thus, although Wisdom presents a less sharply defined concept of miracle than Philo, it is, if anything, even more concerned to insist that the only BNP is God (or a divine agency, wisdom or word). Whereas Philo, who almost certainly shared this view, could in one or two places make it appear that Moses was also a BNP (although I have argued that this cannot be what he meant), Wisdom never creates this impression.

d. The Function of the Miraculous in the Wisdom of Solomon

Overall, the main functions of miracle in Wisdom are to punish the wicked and to reward or protect the righteous, the latter generally being Israel and the former its enemies. We have seen in both Philo and Josephus how these two functions can sometimes be two sides of the same coin, but Wisdom takes this a stage further by insisting that the means for punishment and reward are the same.

Most of Wisdom’s views on the purposes of miracle can be found in Wis. 11:1–12:2. Verses 11:1–14 deal with water featuring both as an unexpected supply from the rock for the Israelites (11:4) and as a river rendered undrinkable through being changed into blood for the Egyptians (11:6), which was in turn poetic justice for the Egyptians’ command to drown Israelite children in the river (11:7). The point about the use of the same element for both purposes is spelt out explicitly: ‘For through the very things by which their enemies were punished, they themselves received benefit in their need’ (11:5). The point about the punishment fitting the crime is repeated at 11:15, which states that the Egyptians were punished by plagues of irrational animals for the crime of worshipping irrational animals. This excursus on punishment goes on to indicate that God’s punishments are nevertheless tempered by both his justice (11:20) and his mercy (11:23). This leads into a statement of one of the subsidiary reasons for punishment, to warn sinners so that they may turn from wickedness (12:2). A further, related reason is to discipline the righteous and give them a taste of what the ungodly suffer, so that the righteous may know of God’s punishment (11:9). Nonetheless, the punishment of the wicked is far more severe (11:14).

Further examples of the same means being used to punish the wicked and reward or protect the righteous are not hard to find: Wisdom’s entire account of the exodus is structured round this contrast (or syncrisis). Thus, for example, at 16:1–4 the creatures that torment the Egyptians are contrasted with the quails that provide food for the Israelites. This version of events, in line with Wisdom’s entire presentation of the exodus, lacks any condemnation of the Israelites for grumbling, but again makes the point that prior to receiving the gift of quails the Israelites had to suffer hunger a short time to show them ‘how their enemies were being tormented’ (16:4). Verses 16:5–14 contrast the locusts and flies whose bite killed the Egyptians with the serpents whose bite did not destroy the Israelites, but merely warned and instructed them. Verses 16:15–29 describe how rail, fire and lightning destroyed the crops of the Egyptians (but not the creatures that plagued them), whereas the Israelites were fed with manna from heaven. Verses 17:1–21 detail the terrors conjured up in the impenetrable darkness that surrounded the Egyptians alone, magnified by their frightened imaginations, while the Israelites enjoyed the light (18:1–4). Here Wisdom seems to have fused the notion that while the Egyptians were in darkness the Israelites enjoyed full daylight (cf. Exod. 10:23; Vit. Mos. 1.145) with the illumination provided by the pillar of fire that guided the Israelites through the wilderness by night (Wis. 18:3). Wisdom 18:5–19 narrates the death of the Egyptian first-born as a fitting punishment for the Egyptian decision to slay Hebrew infants (18:5). This is contrasted with a plague that came upon the righteous Israelites in the wilderness but was quickly stopped by Aaron’s intercession (18:20–25). Why the righteous should suffer this plague in the first place is left unclear (Wisdom makes no reference to any preceding rebellion). Presumably the point is that the righteous once again have to have some experience of the punishment inflicted on the wicked (18:20). The wording of 18:5 (‘and thou didst destroy them altogether by a mighty flood’) might lead the reader to expect the punishment for drowning Hebrew infants to be, not the death of the first-born, but the drowning of the Egyptian army in the Red Sea. This expectation is in fact met at 19:1–5, which describes the Egyptians’ foolish decision to pursue the Israelites, so that they met a ‘strange death’ (19:5); in contrast creation refashioned itself at God’s command to provide a safe way through the sea for the Israelites (19:6–9). The scheme is at times a little forced, but one can see what Wisdom is trying to do.

As we have seen, Philo also talks about the same elements punishing the Egyptians and protecting the Israelites (see Vit. Mos. 1.143–44 and Chapter 3, Section 2 above), but his use of this motif is different from that of Wisdom. For Philo it is evidence of the strangeness of the miraculous: for Wisdom it exemplifies the principle of God employing creation to punish the wicked and reward the righteous.

A closer similarity between Philo and Wisdom is their common insistence on God’s forbearance in punishment (see Chapter 3, Section 4 above). For example, both Wis. 11:15–12:2 and Vit. Mos. 1.110 discuss why God chose to punish the Egyptians through such insignificant creatures as insects, and both conclude that God wished to admonish the Egyptians rather than destroy them. The theme of divine forbearance continues throughout Wisdom 12. Wisdom 12:3–11 discusses why the Canaanites were destroyed only gradually (12:8, 10) in order to give them a chance to repent (12:10), though their stubborn wickedness prevented their making use of this opportunity (12:10–11). Verses 12:12–18 insist that God’s judgments are both righteous and coloured by mildness and forbearance (12:18). Verses 12:19–27 contrast the light chastening of God’s people with the heavy scourging that falls on their enemies (12:22), while insisting that, at least in the initial stage, the punishment always has a didactic or warning function, again to promote repentance (12:20, 26). This emphasis on divine forbearance in Wisdom is distinctly double-edged, however, for by insisting that the Canaanites and Egyptians were given every chance to repent, Wisdom thereby asserts their obdurate wickedness and so justifies their punishment and destruction. Where Wisdom indicates that the Israelites also experienced a light form of didactic or admonitory punishment (e.g. Wis. 11:8–10; 16:10–11; 18:20–25), the implication is that they did respond appropriately (since on each occasion they suffered no great or lasting calamity). Wisdom’s theory of didactic punishment thus serves to underline the righteousness of the Israelites as much as the wickedness of their enemies.

The didactic function of miracles in Wisdom always falls within the wider punishment and reward/rescue functions. Moreover, although some punishment miracles may have this didactic function, Wisdom lacks any indication of a separate, purely evidential function for miracle, which features so strongly in Philo. On the other hand, unlike Philo, Wisdom does suggest an eschatological, or at least future salvation-historical, role for miracle, notably at 5:17–23 and 19:22. It will, however, be convenient to defer further discussion of this to the end of the present section.

The types of miracle in Wisdom include both nature miracles and healing miracles. Indeed, given Wisdom’s insistence on the way miracles are wrought through God’s creation, the term ‘nature miracle’ is for once reasonably appropriate. As we have seen, several of the nature miracles in Wisdom involve plagues of animals or insects or freak storms or disease. These are less miracles in the modern sense of hard anomalies than intensifications of natural phenomena. As is often the case in Josephus, they become miraculous as much through their falling on the appropriate people as through being strikingly odd events. To be sure, there are references to occurrences that would be less easy to explain, such as the Nile becoming contaminated with blood (11:6), or the darkness that afflicted the Egyptians but not the Israelites (17:1–18:4). Even if in the latter case the accompanying terrors are psychologized, the differential darkness remains odd. But it is in the positive miracles of provision and rescue that Wisdom’s nature miracles really come into their own.

First, it is not enough for Wisdom that the manna provided sustenance in the desert. It was not simply ‘bread from heaven’ (though it was that, 16:20), it was ‘the food of angels’, produced without human toil (16:20), that transformed itself to suit every individual’s taste (16:21, 25). This again served a didactic function: to show that man is fed not by crops but by the word of God (Wis. 16:26; cf. Deut. 8:3). Wisdom 16:24–26 may also be making the point that God, who can make all the elements serve him, on this occasion had food produced from the air rather than the ground (cf. Vit. Mos. 1.201–202). For Philo, the most wonderful thing about the manna was that it, in accordance with God’s word, it could not be retained for use on the following day unless that day was a sabbath (Vit. Mos. 1.203; 2.226; see Chapter 3, Section 2 above). Wisdom instead draws the lesson, not stated in the biblical account of the manna, that ‘one must rise before the sun to give thee thanks’ (16:28), since otherwise the manna melted in the heat of the sun (Wis. 16:27; cf. Exod. 16:21). A further peculiarity of the manna according to Wisdom, however, was that though it was readily melted by the sun, fire failed to destroy it (Wis. 16:27; 19:21). The ‘fire’ in question might be the lightning that, along with rain and hail, destroyed the Egyptians’ crops (16:19; 19:20); if so Wisdom has taken considerable liberties with the biblical account by suggesting that the manna was available to feed the Israelites at the same time as this plague took place, that is, while they were still in Egypt. But perhaps the fire referred to is simply that used to cook the manna.40

Wisdom takes no such liberty with the provision of quails, which appear for the Israelites only after the plagues are over (19:11). Like Philo at Vit. Mos. 1.209, Wisdom stresses the bountiful nature of this provision. The quails are a ‘new kind of bird’ (rsv) or perhaps rather ‘new production of birds’ (νέαν γένεσιν ὀρνέων—lxx) that appeared for the Israelites when they asked for ‘luxurious food’ (19:11), and a delicacy (or ‘strange taste food’—ξένην γεῦσιν τροφὴν, hence neb ‘unwonted food’) to satisfy their appetite (16:2–3). In keeping with Wisdom’s idealized portrait of Israel, there is no hint of the condemnation that appears at Num. 11:18–23 in connexion with the people’s desire for meat.

The main deliverance nature-miracle is the crossing of the Red Sea, the two accounts of which effectively frame the long section on the exodus. It is first narrated briefly at Wis. 10:18–19, where wisdom leads the Israelites over the Red Sea but drowns their enemies. The description of the deliverance element is elaborated at 19:6–8, which emphasizes in particular the appearance of dry land in place of the water (19:7). The previous section, 19:1–5 does not elaborate on the drowning of the Egyptians, which is only alluded to, but rather justifies it on the grounds of their obdurate folly and wickedness. In one way, Wisdom heightens the miraculous element of this scene by remarking, ‘the whole creation in its nature was fashioned anew, complying with thy commands’ (19:6) and ‘the elements changed places with one another’ (19:18), emphasizing that something really dramatic is going on, as befits the climax of this recollection of exodus events; the Red Sea crossing thus becomes the prime example of how creation serves its Creator in fighting for his people and against their enemies. On the other hand, as has been discussed above, this language also serves to prevent even this most spectacular of miracles from becoming a hard anomaly: God does not violate the order of his creation, he simply turns it to unusual and dramatic effect.

Unlike Philo, who reports no miracles of healing, Wisdom extracts two healing miracles from the exodus account. The first concerns the plague of serpents in the wilderness (Wis. 16:5–14; cf. Num. 21:4–9). Wisdom characteristically makes no reference to the Israelites’ complaint that led to this punishment. It also alludes only indirectly to the bronze serpent as a means of cure (it would hardly suit its polemic against animal idolatry to be too explicit here). The ‘symbol of … salvation’ (Winston’s translation; lxx σύμβολον … σωτηρίας) at 16:6 presumably refers to this serpent, but the text goes on to stress that healing came, not from looking at this symbol, but from God (16:7). The healing is again reported as the direct act of God at 16:10. Interestingly, Wisdom then goes on to say, ‘For neither herb nor poultice cured them, but it was thy word (λόγος), O Lord, which heals all men’ (16:12). Here Wisdom appears flatly to contradict the view of healing expressed by Philo (and Ben Sira) by insisting that healing is properly the direct action of God without the intermediate aid of human medical techniques. Whether one can press the text further to make the λόγος of 16:12 a divine hypostasis acting to heal is another matter. Most likely the wording here is simply a reflection of Ps. 107:20.

The second healing miracle is described at Wis. 18:20–25. This is based loosely on the account at Num. 16:41–50, where, following the spectacular quelling of the rebellions of Korah, Dathan and Abiram, the people murmur against Moses and Aaron for the deaths of so many Israelites, and are consequently afflicted with a plague that kills 14, 700 of them. The plague is only halted when Aaron runs among the people with his censer and makes atonement for them. Wisdom makes no mention of these rebellions or the ensuing complaints, but starts the account by simply stating that the plague took place (18:20). Wisdom then makes rather more of Aaron’s role. As in the biblical account he makes propitiation through incense (18:21), but he also subdues the punisher (τὸν κολάζοντα, perhaps now conceived as an agent independent of God) both by his word (18:22) and by the cosmic symbolism of his priestly robes (18:24–25). In all this, Aaron shows himself to be God’s servant (18:21). If this account is read on the basis of that in Numbers, where the plague is sent by God as a punishment, it is a little difficult to make sense of it all, since Aaron is then called God’s servant for petitioning God to stop the plague that God himself sent for the express purpose of wiping out his rebellious people. But, as we have seen, Wisdom generally ignores any negative portrayal of Israel in the biblical account. What happens in the present passage may simply be a triumph of rhetoric over logic, but it may be that here Wisdom does not want the reader to attribute the plague solely to divine punishment or correction. The reference to ‘the destroyer’ (ὁ ὀλεθρεύων) at 18:25 perhaps implies some demonic agency opposed to God. Compare the ‘destroying angel’ (lxx τὸν ὀλεθρεύοντα) at Exod. 12:23 (cf. also 1 Chron. 21:15). To be sure, Exodus 11 suggests that the death of the first-born is the work of God, but elsewhere the role of the ‘destroying angel’ can become quite ambivalent. At Jub. 49.2, for example, he is equated with the arch-demon Mastema (see Chapter 6, Section 3 below). But can Wisdom be taken this way?

Wisdom nowhere directly refers to evil spirits, demon possession or exorcism, but there is one further passage that may be relevant in this regard. At 7:20 the speaker, presumably meant to be Solomon, says that wisdom has given him knowledge of, inter alia, πνευμάτων βίας καὶ διαλογισμοὺς ἀνθρώπων. Rsv renders this ‘the powers of spirits and the reasonings of men’, but in a footnote suggests ‘winds’ as an alternative to ‘spirits’ (cf. neb ‘the violent force of winds and the thoughts of men’). The word πνεῦμα occurs some 20 times in Wisdom. Although it sometimes mean ‘wind’ (e.g. 5:23; 17:17 lxx = 17:18 rsv), it much more frequently means ‘spirit’ (e.g. 1:5, 6, 7; 2:3; 5:3; 7:7, 22, 23; 9:17; 12:1; 15:11; 16:14), and it must be taken as doing so at 7:20. The entire verse, in the translation of the rsv, runs

the nature of animals and the tempers of wild beasts,

the powers of spirits and the reasonings of men,

the varieties of plants and the virtues of roots;

In the first and third lines there is clearly a reasonably close synonymous parallelism between the first and second items mentioned. This must almost certainly hold for the second line as well, in which case ‘the violence of winds’ is too much of a mismatch with ‘the reasonings of men’ to be plausible. This at least opens up the possibility that the spirits referred to are evil spirits or demons, given that Jewish tradition elsewhere credits Solomon with power over demons and specialist knowledge of exorcistic techniques (as, for example, in Josephus’s account of Eleazar’s exorcism).

On balance, however, this is not the most probable reading. The first line of the three deals entirely with animals, and the third entirely with plants. It would thus best fit the structure of the verse if the ‘spirits’ referred to in the second line were human ones. Such a reading is supported by the reasonably frequent use of πνεῦμα to refer to human spirits elsewhere in Wisdom (2:3; 5:3; 15:11, 16; 16:14; i.e. in one quarter of the instances). The phrase ‘the violence of spirits and the reasonings of men’ could then refer to the emotional and rational parts of the human soul, distinguishing humans from the wild beasts of the previous line (note that πνεῦμα is also used in connexion with strong emotion at 5:3). Alternatively, ‘the powers of spirits’ may simply be another way of referring to human reasoning powers. Either way there is no reference to demons or to knowledge of exorcistic techniques here. This remains the case even if it is felt that the second line refers to rational creatures in general—spiritual and corporeal; the parallel would then consist in the shared rationality of spiritual creatures, angelic or human. To understand the ‘spirits’ as demons would require the parallel to be between the forces of demons and the reasonings of men, but this is too unlike the parallels in the surrounding lines. It must therefore be concluded that, in common with Philo, Wisdom has nothing to say about demons or exorcism.

The final question that needs to be discussed is how seriously Wisdom takes the miracle stories it narrates or alludes to. The answer is not so clear-cut as in the case of Philo and Josephus, not least because Wisdom never addresses the question explicitly as they do; since it is a different kind of work it does not need to. Indeed, both the liberties Wisdom takes with the biblical narratives in order to make them fit its schema, and its technique of narrating biblical history, including the exodus story, without using any proper names tend to shift it away from the historiographical to the typological. The anonymous exodus narrative becomes an illustration of how God punishes and rewards through the same elements, and how creation is made to serve in this manner. This does not entail, however, that the miracles narrated are thereby dismissed as unhistorical. On the contrary, the rhetorical force of the composition presumably depended in part not only on the audience’s recognition of the events alluded to but on their acceptance of them as part of the factual history of their nation. But this point is not conclusive. One could imagine, for example, that Alexandrian Jews concerned about their status might well derive satisfaction from stories denigrating the Egyptians even if they knew the stories to be fictional.

What lends support to the notion that Wisdom intended the miracle stories to be taken seriously is the manner in which they serve as archetypes for God’s action throughout his people’s history, and, by implication, his people’s future. Thus, as we have already noted, the concluding verse is not simply a conventional doxology, though it also serves a doxological purpose, for it emphasizes that God has ‘not neglected to help [his people] at all times and in all places’. The clear implication is surely that he will continue to do so. Moreover, as we have again already noted, the detailed account of how the elements fought for the righteous Israelites at the time of the exodus is prefigured in the passage at Wis. 5:17–23. But though the allusions in this passage correspond in many ways to the exodus events to be described later, the verbs are all in the future tense. This seems to be a way of indicating that the great saving events of the exodus constitute a pattern for what God will do again in the future. This can only be taken seriously as a promise of future deliverance if the exodus miracles are taken seriously. Whether such future deliverance should be described as ‘eschatological’ in the sense of ‘final’ is another matter. J.P.M. Sweet may well be right that in Wisdom miracles ‘disclose the general principles on which the world is based rather than indicate the presence of the world to come or foreshadow its breaking in. They do not threaten the structure of the world but reveal it’.

Ben Sira

The Wisdom of Yeshua ben Sira was originally written in Hebrew some time around 190–180 bce and subsequently translated into Greek by the author’s grandson. There appear to have been two principal Hebrew recensions underlying two main Greek translations, the second varying from the first mainly by expansion.53 The work survives in its entirety only in Greek (and subsequent translations), and in the Syriac Peshitta, although there are also extant Hebrew fragments covering about two-thirds of the text. Although Ben Sira’s grandson was reasonably faithful to his grandfather’s work, there are places where his translation is not precise. But it could be argued that the complete lxx Greek text represents a form of the work acceptable to some Jewish constituency at, say, the end of the second century bce and is thus equally relevant to the present study. We shall accordingly deal with two forms of the text: the lxx (represented in English translation by the rsv) and the alternative translation (based as far as possible on the Hebrew) of Patrick Skehan. The former will be cited as Sirach and the latter as Ben Sira.

Ben Sira is often seen as a conservative resisting Hellenizing influences on Judaism. A more nuanced view would be to see him as opposed, not to Hellenism as such, but to any form of modernization that threatened what he saw as the ancestral faith. Ben Sira seems to have been quite willing to welcome Hellenistic ideas, such as popular philosophical commonplaces and ‘scientific’ medicine, that he felt could be fitted in with his Judaism. On the other hand, although Ben Sira resembles Proverbs more than any other biblical wisdom book, he moves beyond Proverbs in the way he incorporates specifically Jewish traditions into his wisdom teaching. This comes about not only through his equation of wisdom and torah and his references to the priesthood and the cult, but most strikingly through his section on the ‘Praise of Famous Men’ in which he rehearses the achievements of some of the most notable figures from biblical history. Ben Sira is principally concerned with passing on good advice to his younger male contemporaries on such diverse topics as women, friendship, riches and poverty, behaviour at banquets, and death. But in his discourse on the role of the physician and his ‘Praise of Famous Men’ he incidentally touches on the matter of miracle.

A convenient place to begin is the discourse on the role of the physician (Sirach 38:1–15). At first sight this is of interest mainly as an extended discussion of healing, which discusses the respective roles of God and the physician, allowing that the latter has his uses while insisting that healing comes ultimately from the former (Sirach 38:2). Although one should ‘give the physician his place, for the Lord created him’ (Sirach 38:12), his role is dependent upon God (Sirach 38:14). Here no ‘miraculous’ healing is in view, but rather healing by natural means, since the passage refers to the ‘skill of the physician’ (Sirach 38:3) and to medicines and the role of the pharmacist (Sirach 38:4, 8; cf. Leg All. 3.178). This passages thus serves to legitimate the medical practitioner while maintaining that God is still the ultimate source of healing. Priority is to be given to the latter, for when one is ill one is first to ‘pray to God, for it is he who heals’ (Ben Sira 38:9) and to repent of sin and offer the appropriate sacrifice (38:10–11); only then should the physician be summoned, and he too is to pray for guidance in the exercise of his skill (38:12–14). Ben Sira thus appears to concur with Philo that healing comes from God, but that God generally chooses to exercise his healing power through the ministrations of a physician.

Two verses in the passage require some further comment, however. Sirach 38:15 expresses what, at least in isolation, would appear to be a highly negative judgment on the medical profession: ‘He who sins before his Maker, may he fall into the care of a physician’. In context such a judgment seems out of place. One way out of the difficulty would be to see the verse as expressing the wish that one who sins should fall ill (and hence need the services of a physician). An alternative is to adopt Skehan’s preferred reading: ‘Whoever is a sinner toward his Maker will be defiant toward the doctor’ (since the doctor’s skills come from God).60 Either way, the verse does not then turn out to contradict the main thrust of the passage.

The second noteworthy verse contains an allusion to one of the Mosaic miracles: ‘Was not water made sweet with a tree in order that his [or its] power might be made known?’ (Sirach 38:5); ‘Was not the water sweetened by a twig that people might learn his power?’ (Ben Sira 38:5). This follows the statement that ‘The Lord created medicines from the earth, and a sensible man will not despise them’ (Sirach 38:4); or ‘God makes the earth yield healing herbs, which the prudent should not neglect’ (Ben Sira 38:4). The piece of wood that sweetened the waters of Marah (Exod. 16:22–25) is thus used as an example of (or at the very least, a parallel to) the healing herbs created by God for the use of prudent men. Ben Sira thereby finds a biblical warrant for the use of medicinal substances. Whether one takes ἰσχὺν αὑτοῦ at Sirach 38:5 to refer to ‘his [i.e. God’s] power’ or ‘its [i.e. the wood’s] power’ the point is basically the same. The piece of wood Moses employed to sweeten the water had been endowed by God with healing power for the purpose, just as (other) herbs have been endowed by God with the power of healing human ailments. The effect is thus not to make the latter type of healing more miraculous by association with the Mosaic story, but to make the Mosaic story less miraculous. God remains the ultimate source of the healing power in both cases, but for Ben Sira, healing ceases to be a striking surprising event.

Ben Sira is also remarkably restrained about the miracles associated with Moses in his section on the ‘Praise of Famous Men’ (chs. 44–50). Indeed, given the interest in Moses exhibited in other texts, the five verses devoted to him at Sirach 45:1–5 seem paltry in relation to what Ben Sira has to say about, for example, Aaron or Joshua. Moreover, the only allusion to miracle in the verses on Moses reads curiously in Sirach 45:3: ‘By his words he caused signs to cease’ (ἐν λόγοις αὑτοῦ σημεῖα κατέπαυσεν). Perhaps this should be understood as a reference to Moses’ successful contest against the Egyptian magicians, or perhaps the point is that several of the plagues were terminated at Moses’ request. Either way, it seems odd to connect Moses only with the cessation of miracles. Skehan’s translation of Ben Sira 45:2–3 gives what is clearly better sense: ‘God made him like the angels in honor, and strengthened him with fearful powers, Wrought swift miracles at his words and sustained him in the king’s presence.’ But even if this translation is correct the reference to the king’s presence would seem to restrict the miracles to those Moses performed before Pharaoh, that is, the signs with rod and hand and the plagues.64 The great miracles of the Red Sea crossing and wilderness wandering are then passed over in silence (apart from the allusion to sweetening the waters of Marah, in another context). Given the miracles Ben Sira recalls in connexion with other figures, this is curious.

Prior to Moses, Ben Sira has dealt (briefly) with the patriarchs Abraham, Isaac and Jacob. Following Moses he goes on to celebrate Aaron, Phinehas, Joshua, Caleb, the Judges (as a group), Samuel, Nathan, David, Solomon, Elijah, Elisha, Hezekiah, Josiah, Jeremiah, Ezekiel, Job, the Twelve Prophets, Jeshua, Zerubbabel and Nehemiah, before rounding off with brief references to Enoch, Joseph, Shem, Seth, Enosh and Adam and concluding with a lengthy praise of the high priest Simon. Many of these figures have few or no miracles associated with them in the biblical narrative, but where they do Ben Sira often focuses on them. Thus Ben Sira 45:18–19 briefly describes the miraculous suppression ofthe rebellions of Dathan, Abiram and Korah against the priestly authority of Aaron. Joshua is remembered, among other things, as the one for whom the sun stood still and whose enemies were vanquished by divinely sent hail (46:4–6). God’s aid in battle is also recalled in connexion with Samuel, in response to whose entreaty the Lord thundered from heaven to confound the Philistines (46:16–18). Elijah and Elisha are remembered as much for their miracles as for anything else. Elijah shut up the heavens, brought down fire three times, revived a dead child, and was finally taken up in a whirlwind to return before the day of the Lord (48:1–10). Elisha was then filled with Elijah’s spirit so that ‘Twice as many signs he wrought, and marvels with every utterance of his mouth’ (Ben Sira 48:12; Sirach lacks this bicolon). Indeed Ben Sira goes on to recall that even Elisha’s corpse wrought wonders: ‘Nothing was beyond his power; from where he lay buried, his dead body prophesied. In life he performed wonders, and after death, marvelous deeds’ (Ben Sira 48:13–14). The reference is to the miraculous revival of corpse thrown into Elisha’s grave (2 Kgs 13:21). It is noteworthy that Ben Sira refers to this port mortem miraculous activity as ‘prophesying’ (ἐπροφήτευσεν, Sirach 48:13). It is similarly to this aspect of Isaiah’s career (in conjunction with Hezekiah’s) that Ben Sira calls attention. When the people called out to God at the time of Sennacherib’s invasion, God ‘saved them through Isaiah’ by smiting the Assyrian camp (Ben Sira 48:20–21); Isaiah similarly lengthened the life of the godly king in conjunction with the sign of the reversing sun (48:23). The only other prophetic activity attributed to Isaiah is his revelation of things to come in the distant future (48:24–25).

Despite this catalogue of miracles, it is clear that they are not the main focus of attention. Miracles are recalled in connexion with only a minority of the figures Ben Sira deals with (Moses, Aaron, Joshua, Samuel, Elijah, Elisha, Hezekiah and Isaiah), and in most of these cases they play only a minor role in the description of the character’s career. It is really only Elijah and Elisha who seem to have been transformed primarily into wonder-workers. In the main, this is readily explicable. Ben Sira has chosen to recall some of the most spectacular and noteworthy incidents of each of his heroes’ careers. By their very nature, miracles are spectacular and thus obvious candidates for inclusion. The cycle of miracle stories connected with Elijah and Elisha would thus naturally stand out as the most striking feature of their lives. The only real oddity is the low key treatment of Moses. Perhaps the reason for this is Ben Sira’s desire to emphasize Aaron as the source of the Levitical priesthood; the only other character who receives as much attention as Aaron is the high priest Simon. Thus although ‘we might have expected more development of Moses’ role … in fact he serves mainly to introduce the much longer eulogy of his ‘brother’ Aaron’.68 The de-emphasis on Moses is thus most probably an accidental by-product. It nevertheless remains noteworthy that a Jewish author can recall the great characters of Israelite history, including many of the great miracles wrought through them, while apparently ignoring the great cluster of miracles that most of the Jewish authors examined here took to be the most important of all.

Given that miracle is not Ben Sira’s main theme, one should be cautious about pressing him for any precise theology of miracle. In particular, he occasionally seems a little careless over whether miracles are to be attributed to men or to God. Skehan’s translation of Ben Sira 45:2–3 makes Moses a PNP, since God is said to have ‘wrought swift miracles at his words’. At Sirach 45:3, however, it appears to be simply Moses’ words that cause signs to cease, without explicit reference to God’s role as BNP (see above). In both versions the destruction of the rebels against Aaron is attributed solely to God (45:18–19). Joshua is presented as a PNP when the Lord sends hail to defeat his enemies in response to his prayer (46:5–6), but the strict sense of the text makes him a BNP in holding back the sun with his own hand (46:4). Here, however, one may surely allow Ben Sira a certain amount of poetic license. Samuel is purely a PNP when God thunders against the Philistines in response to his prayer (46:16–17). In contrast, Elijah appears as a particularly powerful MNP, for although Elijah is said to have shut up the heavens, called down fire three times, and raised a dead child, all these feats were performed by God’s word (Ben Sira 48:3) or will (Ben Sira 48:5; Sirach 48:5 has ‘word’ again). Elisha, on the other hand, comes perilously close to being a BNP (see Ben Sira 48:12–24 cited above)—alive or dead! The only feature of the text that mitigates this conclusion is the notice that Elisha received Elijah’s spirit (48:12), which might be taken as indicating that Elisha’s source of miraculous power was ultimately the same as Elijah’s, namely God. When Sennacherib invades Judah the people act as joint PNP in calling upon God, who duly destroys the Assyrian army (48:18–21). One curious feature of this account is that God is said to have acted through Isaiah (48:20), although Isaiah is left with no clear role to play since it is God (Ben Sira 48:21) or his angel (Sirach 48:21) who smites the Assyrian camp. Finally, Isaiah is treated as a BNP when, according to Sirach 48:23, ‘In his days the sun went backward, and he lengthened the life of the king.’

How one evaluates this data depends in no small part on what one takes the function of the ‘Praise of Famous Men’ section to be. If it is seen as a praise of God in human obedience following the praise of God in nature (42:15–43:33), then all the miracles narrated can be seen within the framework of God’s acts in history. But it seems more natural to see the ‘Praise of Famous Men’ as some form of encomium on the great figures of biblical history.72 It may be that this thereby demonstrates that wisdom has a special heritage in Israel, but the impression given is that Ben Sira ‘does not praise so much God’s action in them and through them as the men themselves, whom he sees as giants, examples often worth propounding and imitating by the wise man’.74 This does not necessarily mean, however, that Ben Sira deliberately intends to portray any of Israel’s heroes as BNPs. If Ben Sira is praising not so much God as these men, then it is noteworthy how often he in fact makes the men PNPs (Moses, Joshua, Samuel) or MNPs (Elijah, Isaiah), at least some of the time, thereby emphasizing that it is God who acts as BNP. It is only Elisha who forms a total exception here. Again, given that the passage is poetry, one must allow for a certain amount of poetic license in the way in which its subjects are praised. The restraint that makes Ben Sira’s heroes PNPs or MNPs most of the time may well be a more reliable index of his views than the occasional license that makes some of them appear as immanent BNPs.

Of more significance than Ben Sira’s assignment of miracles to God or men is his association of miracles with so many great figures of Israel’s past, especially prophets. Here is an indication from a conservative Jewish writer of the second century bce that great men of God—especially prophets—might be expected to perform miracles (or to have miracles performed for them). Ben Sira does not discuss the legitimation of a prophet, and thus cannot be cited as direct evidence that a would-be prophet would be expected to legitimate himself with miracles. But he does show that prophets and miracles could be quite closely associated in the Jewish mind, even to the extent that the miracles could come to be seen as the most important activity of the prophet, and this may well form a significant part of the context for understanding Jesus’ miracles. In this connexion it is also of interest that Ben Sira can place some emphasis on the healing (and resuscitation) miracles performed by Elijah, Elisha and Isaiah while placing very little emphasis on the spectacular miracles of deliverance associated with Moses.

This last finding is in some contrast with the one passage in Ben Sira that does look back to the saving events of the exodus in the hope of some form of eschatological deliverance (Ben Sira 36:1–22; Sirach 36:1–17). Although there is no explicit reference to the exodus events, the language of the prayer clearly evokes the language associated with the exodus: ‘Give new signs and work new wonders; show forth the splendour of your right hand and arm; Rouse your anger, pour out wrath, humble the enemy, scatter the foe’ (Ben Sira 26:6–9). Scholarly opinion is divided over whether this poem should be attributed to Ben Sira, was taken over by him,77 or is a subsequent addition to his work. Of these three possibilities the last seems most likely. Ben Sira shows little interest in eschatology elsewhere (apart from the reference to the return of Elijah, which is different in character and may also be secondary), and nowhere else refers back to the exodus as a paradigm saving event. Neither does the rest of the book show much sign of the kind of political crisis that would lead to a desire for the destruction of oppressors (Ben Sira 36:11), the smashing of hostile rulers’ heads (36:12), or the raising of the divine hand against foreigners (36:3). Such language would much better fit the time of the Maccabean crisis. Unless one is to suppose that Ben Sira lived on into the reign of Antiochus IV and added this passage in response to the Hellenizing crisis, it is most likely to be an insertion into his work. To be sure it picks up on the theme of the description of the God of justice in the preceding verses (Ben Sira 35:14–26), but there is a change of genre from description to petition. The preceding poem provides a convenient insertion point for a later interpolator, but its point is to urge the individual to just conduct (e.g. ‘keep the Law’, 35:1; ‘refrain from evil’, 35:5; ‘pay your tithes’, 35:11; ‘offer no bribes’, 35:14), not to work up to a condemnation of the Seleucid overlords. The poem in 36:1–22 is thus probably not evidence for the thought of Ben Sira; what it does show is that in time of political crisis some second-century bce Jew found it natural to turn to the exodus as the paradigm for a hoped-for deliverance in the near future. This is not in itself surprising, but it does provide further evidence for what we saw much more clearly in the Wisdom of Solomon.


One might not expect Jewish wisdom literature to cast a great deal of light on the miracles of the historical Jesus, though there is no a priori reason why it should not have influenced the literary presentation of those miracles in the Gospels. On the face of it, the Wisdom of Solomon, like Philo, is remote both from the historical Jesus and from his portrayal in the Gospels. Wisdom is slightly closer than Philo to Jesus and the Gospels in showing some interest in healing miracles, although, like Philo, it shows no explicit interest in demons, possession or exorcism. Wisdom may also have some affinity with the Fourth Gospel in that the former’s portrayal of Wisdom as the agent empowering Moses’ miracles could be seen as vaguely similar to the latter’s Logos Christology. It is conceivable that the author of Wisdom was a direct contemporary of Jesus. If so, it is noteworthy that his ideas on eschatology seem very different. To be sure, much of Wisdom’s eschatology focuses on the vindication of the just in the afterlife, but in its this-worldly aspect, while it can indeed attach eschatological (or, at least, future salvation-historical) significance to miracles, the miracles in question are envisaged as being like those that attended the exodus: God employing nature to spectacular effect to fight for his people against their foes. This conception seems closer to that of the sign prophets as reported by Josephus (on whom see Chapter 11 below) than to Jesus as described in the Gospels.

For the purposes of this study, the most significant point to emerge from the discussion of Ben Sira is that his survey of Israelite history makes a strong association between prophets and miracles while making very little of the miracles associated with Moses. Since there is no reason to suppose that Ben Sira is particularly interested in miracle per se, this could be evidence that Palestinian Jews might automatically make a connexion between miracles and prophets even if the prophetic candidate performed no spectacular miracles of the exodus type (as many of the ‘sign prophets’ apparently promised to). Moreover, since Ben Sira’s survey includes many of the healing and resuscitation miracles associated with Elijah, Elisha and Isaiah, it might well be natural for someone who performed healing miracles to be seen as a prophet in that tradition (cf. Mk 6:15; 8:28; Lk. 4:24, 27; 7:16, 39; 13:33; 24:19).

The move from the Wisdom of Solomon to Ben Sira is also one from Hellenistic Jewish to Palestinian Jewish literature. The next chapter will examine another example of what may have been middle-of-the-road Palestinian Jewish literature: the Liber Antiquitatum Biblicarum of Pseudo-Philo.

Chapter 5

Miracle in Pseudo-Philo


Although Pseudo-Philo did not leave such an extensive corpus as either Josephus or Philo of Alexandria, his Liber Antiquitatum Biblicarum (henceforth LAB), narrating the biblical story down to the death of Saul, is both substantial enough and sufficiently comparable with parts of Josephus’s Antiquities and its parallels in Philo to be worth treating in the same way. The identity of Pseudo-Philo is unknown (the LAB was once erroneously believed to be the work of Philo of Alexandria, with texts of whose work it was circulated, hence the designation ‘Pseudo-Philo’). He (or she) appears to have been a Palestinian Jew writing in Hebrew toward the end of the first century ce, although some scholars argue for a pre-70 date. It is possible, then, that Pseudo-Philo was a Palestinian contemporary of Jesus, but rather more likely that he was a contemporary of the Evangelists. Even in the latter case he may well have been drawing on traditions that were current in Palestine in the time of Jesus. Many scholars see him as a representative of a form of mainstream non-sectarian Judaism.5 If this is correct then Pseudo-Philo should provide some valuable clues about what significance many first-century Palestinian Jews attached to the miraculous, at least in relation to the biblical tradition. On the other hand, it provides little or no explicit reflection on the miraculous, since the author, unlike Philo and Josephus, shows no consciousness of needing to justify Jewish traditions to a possibly sceptical Hellenistic audience. Instead Pseudo-Philo appears to be a Jew encouraging other Jews on the basis of shared traditions and beliefs, which happened to include some of those about God’s miraculous interventions in history.

The Concept and Language of the Miraculous in Pseudo Philo

Pseudo-Philo’ s Book of Biblical Antiquities survives only as a Latin text (in several manuscripts) that appears to be a translation from a Greek version of a Hebrew original. One might therefore suppose that it was pointless to discuss Pseudo-Philo’s vocabulary of miracle, since only that of the Latin translation survives. On the other hand, the factors that indicate the previous existence of a Greek version and Hebrew original also suggest that both translators worked in a fairly literal fashion and so may have preserved something of the linguistic patterns of the original author. The fact that the Latin vocabulary of miracle in Pseudo-Philo appears to reflect the usage of the Hebrew Bible lends further encouragement to this view.

Pseudo-Philo’s three most common words for miracle occur together at LAB 9.7: ‘et per eum faciam mirabilia in domo Iacob, et faciam per eum signa et prodigia populo meo que non feci ulli’ (‘and I will do wonders in the house of Jacob through him [sc. Moses] and I will perform through him signs and wonders for my people that I have not done for anyone else’).8 The phrase ‘signa et prodigia’ presumably represents an underlying Greek σημεῖα καὶ τέρατα, which would itself represent the biblical אותת ומפתים (e.g. Deut. 6:22). ‘Mirabilia’ could well translate θαυμάσια in turn translating פלאות. Here it will be convenient to refer to the Latin terminology of the surviving texts, in the (reasonable) hope that it consistently reflects the Hebrew terminology of the original.

It is another matter how far this terminology indicates a well defined or unitary concept of miracle in Pseudo-Philo. For one thing, Pseudo-Philo often narrates what we might consider to be a miracle without attaching any particular term to it, whereas the terminology that occurs at LAB 9.7 often occurs elsewhere in summary contexts in which signs or wonders are alluded to in general but not specifically described (e.g. LAB 12.2; 14.4; 20.4; 25.6; 27.7; 30.5; 53.8). For another, Pseudo-Philo lacks the kind of conscious reflection on the miraculous in detailed narrative or apologetic argument that helped define the concept in Philo and Josephus. And for another, Pseudo-Philo applies his miracle terminology to a rather diverse range of phenomena.

Given Pseudo-Philo’s emphasis on God’s control of events (which will be discussed further below), it is at least clear that, for Pseudo-Philo a miracle is an act of God (or occasionally, of an angel who presumably acts as God’s proxy). Those events that would strike us as miraculous tend to be striking interventions by God either to save Israel in general or the righteous in particular, or to inflict an appropriate punishment. Pseudo-Philo shows little interest in the epistemological aspect of such divine deeds, except insofar as Israel tends to forget or disbelieve the great saving acts God has performed for past generations (e.g. LAB 25.6). Whereas Josephus allows a certain epistemological ambiguity to miracles, and Philo makes a point of insisting on their being clear-cut evidence of God’s activity, Pseudo-Philo is more interested in them simply as divine deeds. To be sure, they then have a certain evidential value, since both Israel within the narrative and, even more, the implied reader of the narrative are made aware that God again and again acts in history to save or to punish. Indeed, Pseudo-Philo often uses the verb ostendere (‘show’, ‘display’, ‘disclose’) in connexion with God’s mirabilia (e.g. LAB 26.5; 27.7; 30.5; 28.1); but more often God is simply said to ‘perform’ (facere) mirabilia (LAB 9.7; 12.2; 20.4; 30.7; 51.6), miracula (32.13), and [signa et] prodigia (9.7; 53.8).

Pseudo-Philo seldom calls explicit attention to the oddity or seeming impossibility of the miracles he narrates (there are one or two possible exceptions such as the notice that the miraculous birth of sons to Isaac’s wife in their third year of marriage is something that will not be repeated—LAB 32.3). That he regards them as strikingly surprising events is perhaps indicated by his frequent use of the word mirabilia (‘wonders’) to denote them (14 times: LAB 9.7; 12.2; 14.4; 20.4; 26.5; 27.7 bis; 28.1, 2; 30.5, 7; 32.12; 35.2; 51.6) together with one of miracula (LAB 32.13). In contrast the word prodigia (‘portents’) is used only four times (LAB 9.7; 25.6; 42.5; 53.8). On two of these occasions it is used in the stock phrase signa et prodigia (9.7; 42.5), and in three out of the four it is used in connexion with Moses and the exodus events (9.7; 25.6; 53.8). Its employment thus seems to reflect the biblical pattern of using ‘signs and wonders’ to refer especially to the great saving acts of the exodus. It may or may not be significant that the first time signa et prodigia appears (9.7) it is in conjunction with mirabilia, which could be an indication that Pseudo-Philo regards these terms as virtually synonymous, or at least supposes that ‘signs and portents’ are also ‘wonders’.

As noted above, Pseudo-Philo very rarely incorporates any of these miracle words into a direct narration of a miraculous event. At LAB 26.6 Kenaz (a mere name in the Bible, but a major character in Pseudo-Philo) finds that some illicitly acquired magical stones turn out to be as indestructible as God said they were, and then praises God for his mighty works (virtutes—a word occasionally used in Pseudo-Philo, like the New Testament δυνάμεις, to mean ‘mighty works’ in the sense of ‘miracles’, although virtus in Pseudo-Philo more often means ‘power’ or ‘strength’). At 27.7 Kenaz prays for a wonder (unum de mirabilibus) that is also to be a sign of victory (signum salutis), namely that if on entering the Amorite camp and drawing his sword from his scabbard the sword shines and the Amorites recognize that he is Kenaz, they will be delivered into his hands. His prayer is answered, and Kenaz is not only clothed with the spirit of power (spiritu virtutis) but assisted by a pair of angels to defeat the Amorites (27.10). Moreover, he subsequently has difficulty freeing his hand from his sword, since his hand has taken on its virtus (27.11). Aod the magician promises to show the Israelites the sun at night so that they may know that his (Midianite) gods are powerful (virtutem habentLAB 34.1). At LAB 35.6–7 Gideon requests a sign (signum) from an angel, and this is given in the form of water poured on a rock becoming half fire and half blood. Finally, at LAB 61.5–7, God sends David the angel in charge of power (angelum prepositum super virtutem) to assist his defeat of Goliath. All these examples are additions to the biblical account (in the case of Kenaz and Aod, very substantial additions) and may thus represent Pseudo-Philo composing more freely than when reproducing the substance of the biblical narrative. In these instances there seems to be a tendency for Pseudo-Philo to associate power (virtus) with the miraculous, and there are two examples of miraculous signs. Unfortunately these examples are too sparse to fix the meaning of mirabilia or prodigia for Pseudo-Philo.

One may obtain a clearer idea of the sort of events these words designate by identifying what the various characters who employ them in Pseudo-Philo are most likely to be referring to. The speeches assigned to Deborah are a good starting place, since Pseudo-Philo makes her a reliable spokeswoman for God, and thus a reliable indicator of his own views. At 30.5 Deborah tells the people oppressed by Jabin and Sisera that God ‘showed you not a few wonders (mirabilia); and for your sake he commanded the luminaries, and they stood still in their own places; and when your enemies came against you, he rained down hailstones on them and destroyed them’. This is a reference to the events narrated in Josh. 10:11–14 (though omitted from Pseudo-Philo’s account of Joshua). Two verses later (LAB 30.7) Deborah promises that God is about to work similar wonders (mirabilia) for the people of Israel to defeat their enemies. At 31.1–2 this promise is fulfilled by the stars going forth and burning up over eight million men of Sisera’s army. The biblical song of Deborah that follows Sisera’s defeat (Judg. 5) is largely confined to celebrating this one victory. In Pseudo-Philo’s version (LAB 32), however, Deborah is made to recall the history of God’s acts for his people starting with the rescue of Abraham from the furnace (another of Pseudo-Philo’s non-biblical additions, on which see below). At 32.12 she declares, ‘Therefore we do not cease singing praise, nor will our mouths be silent in telling his wonders (mirabilia)”; and at 32.13 she has the Israelite ancestors declare, ‘The Lord has not forgotten the least of the promises he established with us, saying, “Many wonders (miracula) will I do for your children.” ‘. This strongly suggests that the divine actions related in the intervening verses are to be considered examples of mirabilia and miracula.

These actions include not only Abraham’s rescue from the furnace, but also the gift of a son to Abraham in his old age through a barren wife, the attempted sacrifice of Isaac (and God’s last-minute intervention to prevent it), the birth of two sons to Isaac through a barren wife, the liberation from Egypt (described very briefly), the law-giving at Sinai accompanied by a theophany of cosmic proportions (described at rather more length), the sun and the moon standing still for Joshua, and the stars’ aid in defeating Sisera (32.1–11). Mirabilia thus appear to include striking divine military interventions, miraculous births, and miraculous rescues.

Mirabilia, signa and prodigia are often associated with Moses and the exodus. Thus, as we have already seen, God promises to work all three through Amram’s as yet unborn son (LAB 9.7). God then tells Miriam that he will work signs (signa) through her future brother (LAB 9.10); in the same breath he tells her that he will save his people through him and that the waters will be dried up through him. While Moses is up the mountain, the people complain to Aaron, ‘that Moses through whom wonders (mirabilia) were done in our presence has been taken from us’ (LAB 12.2). At LAB 19.11 God refers to Moses’ staff through which he worked signs (signa). The people at the time of Kenaz confess to God that they ‘are found not to believe in the wonders (non credentes prodigiis) that you did for our fathers from the time you brought them out of Egypt until this day’ (LAB 25.6). God tells the boy Samuel that

I revealed myself to the house of Egypt and chose for myself a prophet Moses my servant and did wonders (prodigia) through him for my people and took vengeance on my enemies as I wished. I led my people in the wilderness and enlightened them, as they saw (LAB 53.8).

Strictly speaking these mirabilia, signa and prodigia are never enumerated, but it is clear that what Pseudo-Philo has in mind must include the Red Sea crossing and the plagues, if not also the manna, the well of Marah, and the theophany at Sinai. Pseudo-Philo never explicitly refers to Moses’ demonstration before Pharaoh, but he often appears to assume that his audience will be familiar with more of the biblical narrative than he actually narrates, so this may also be in view when he talks about the signs worked through Moses’ staff.

There are one or two other instances in which the reference of the miracle-word seems reasonably clear. Thus, for example, when at LAB 28.2 Kenaz refers to ‘all the wonders (mirabilia) that came upon those who sinned’ he can only be referring to the discovery of the sinners by lot and their subsequent punishment by being burnt to death at God’s command (cf. 26.5). Here there is nothing more miraculous than the lot and the subsequent use of Urim and Thummim to determine the appropriate punishment, since the text indicates that the fire was a purely natural one into which Kenaz had the sinners thrown (26.5). When Gideon asks the angel of the Lord where the wonders (mirabilia) are that he has heard about, he presumably has in mind God’s stunning military interventions against Israel’s enemies, since he at once goes on to lament Israel’s subjection to the Midianites (LAB 35.2). On the other hand, when Hannah rejoices in the wonders (in mirabilibus) that God has done for her, she is presumably referring to the birth of Samuel, despite her former barrenness (LAB 51.6). She goes on to encourage her husband Elkanah to sing praises about the Lord’s wonders (signa) also. Manoah asks whether he is not ‘worthy to hear the signs and wonders (signa et prodigia) that God has done among us or to see the face of his messenger’ (LAB 42.5), but in this case all he has missed is the annunciation by an angel to his wife that she will bear a son (and he does not believe her account in any case).

At first sight this seems to suggest that a wide range of phenomena can be regarded as miraculous by Pseudo-Philo. On closer examination, however, what links them is not so much how surprising or supernormal they are, but how significant they are in their effects. The children born to Hannah and Eluma (Samuel and Samson) are destined to play important roles in Israel’s affairs. The removal of the sinners from Israel at the time of Kenaz allows this first great judge to win a substantial victory against the Amorites (LAB 27). The other events described as miracles, the births of Isaac and Jacob, the survival of Abraham, and the great events of the exodus, together with the military victories under Joshua and Deborah, are also of some moment in Israel’s salvation history. It thus appears that most of the events Pseudo-Philo labels as miraculous would fall under the definition, ‘striking divine interventions that have significant consequences for the survival and welfare of God’s covenant people’.

Thus far, I have treated mirabilia, prodigia and signa as if they were virtually synonymous for Pseudo Philo. This seems to be largely the case in most of the examples treated so far, but Pseudo-Philo can also use signum in a more distinctive sense, which must now be examined.

When signum is not simply used as another word for miracle, as in the stock phrase ‘signs and wonders’, it often means ‘sign’ in the sense of an event that reveals God’s will to the enquirer. Such signs may be striking, as in the water that becomes both blood and fire for Gideon (LAB 35.7) or Kenaz’s blazing sword (LAB 27.7), but they can also be more mundane. When Jael is planning to kill Sisera, she asks for two signs that God has heard her and that she will succeed: the first is that he will ask her for a drink of water immediately on waking (LAB 31.5), and the second is that when she returns armed with a stake, Sisera will not wake up when she throws him out of bed (31.7). When David kills a bear and a lion with stones, this is a sign to him that he will subsequently kill Goliath with stones (59.5). For the signs that are to befall Saul in confirmation of what Samuel has told him, the reader is simply referred to the biblical Book of Samuel (LAB 56.7). More exotic is the sign Eli predicts will hold from then on, that if someone is mysteriously called twice it is the doing of an evil spirit, but if three times, then it is an angel calling (LAB 53.4). When used in this sense of a confirmatory sign, then, a signum may be miraculous, but it need not be.

The Distribution of Miracles in Pseudo-Philo

Since Pseudo-Philo retells the biblical story from Adam to the death of Saul, one might expect the distribution of miracles in his account broadly to reflect that of the underlying biblical narrative. This is correct only as a first approximation, however. For one thing, not all parts of the biblical narrative receive equal attention from Pseudo-Philo. Apart from the stories of Noah and Abraham, the book of Genesis is passed over in the most summary fashion. The story of Moses and the exodus is dealt with in more detail, but even then Pseudo-Philo seems to assume that his audience will know the story already, so much of it is omitted (including, for example, Moses at the burning bush and performing signs before Pharaoh) and other parts of it are told in the most summary form (for example the ten plagues, or rather nine of them, are briefly summarized at LAB 10.1). It is the Book of Judges that subsequently receives the greatest attention.

For another thing, Pseudo-Philo not only omits and summarizes material found in the Bible, he also expands on selected incidents and creates more of his own. Thus, for example, Abraham is made the subject of a saving miracle when his story is joined to that of the tower of Babel through a story drawn partly from that of the fiery furnace of Daniel 3 (LAB 6). The topos of those faithful to God being miraculously saved from a fiery punishment that is then turned against their wicked persecutors is repeated at LAB 38.3–4. Here the persecutor is the biblically innocuous Jair (Judg. 10:3–5), who is turned by Pseudo-Philo into an idolatrous promoter of Baal worship. Pseudo-Philo’s creation of the character of the ideal judge Kenaz provides an opportunity for the addition of a number of miraculous, or at least bizarre, incidents that have no obvious foundation in the biblical text (LAB 25.12; 26.6, 8; 27.7–11), although the story of Kenaz’s military success clearly draws from the biblical account of Gideon.

A further complication is that Pseudo-Philo often has characters refer back to incidents, including miraculous ones, that were not actually narrated in their place. An example of this already alluded to is Deborah’s celebration of divine interventions at the time of Joshua (LAB 30.5; 32.10). Finally, as we have seen, Pseudo-Philo does not have an entirely clear-cut concept of ‘miracle’, and in any case it is not always obvious what one should count as a separate miraculous incident.

As a second approximation, one might say that Pseudo-Philo narrates or refers to three miracles in the patriarchal period: Abraham’s rescue from the furnace (LAB 6.15–18) and the miraculous births of Isaac (8.1–3; 23.8; 32.1) and Isaac’s sons (32.5). Roughly 11 miracles are associated with the exodus-wilderness period: the set of plagues (LAB 10.1), the crossing of the Red Sea (10.5–6), the stunning theophany at the law-giving on Mt Sinai (11.4–5), the branch that sweetened the waters of Marah, which well subsequently followed the Israelites about in the wilderness (10.7; 11.15; 20.8), the manna and quails (10.7; 19.10; 20.8); the transformation of Moses’ appearance on descending from the mountain (12.1), the writing disappearing from the tablets (12.5), the effects of the ordeal by drinking water containing the dust of the golden calf (12.7), the punishment of Korah and his fellow rebels (16.1–7), the blossoming of Aaron’s rod (17.1–4), and the pillars of fire and cloud (10.7; 20.8). It will be apparent, however, that these are not all of equal magnitude, and that they hardly all fall into the same category. Joshua has two miracles: the luminaries standing still and the fall of hail that strikes his enemies (3.5; 32.10). Kenaz experiences perhaps three miracles: the erasure of the magical books and the replacement of the precious stones (26.8), his shining sword and single-handed victory in battle, aided by angels (27.7–9), and his inability to release his sword without more Amorite blood (27.11). To this one might add the casting of lots to unmask the sinners and the use of Urim and Thummim to determine their punishment (25.3–5), since the text later refers back to these events as mirabilia (28.2). Other miracles occurring in the period of the Judges include the stars burning Sisera’s army (31.2; 32.11), the sign performed for Gideon (35.6–7), the rescue of the faithful men punished by the idolatrous Jair (38.3–4), the miraculous births of Samson and Samuel (42.3, 51.1), the ignition of Manoah’s sacrifice (42.9), the punishment of the idolatrous Micah (44.9; 47.12), the repeated prostration of the Philistine god Dagon before the ark (55.3) and the ark’s return (55.6–8); this forms a total of, say, 12 miracles to which may be added the magic performed by Aod (34.1–5) and associated with the stones and books discovered by Kenaz’s search (25.12; 26.6). There are even four or five further miracles associated with David and Saul. David sings a song to exorcise Saul’s evil spirit (LAB 60); his defeat of Goliath is aided by an angel (61.5–8); and after Goliath is slain God changes David’s appearance so that no one can recognize him (61.9). Finally, the shade of Samuel makes it clear that it is at God’s command, not that of the medium of Endor, that he has appeared (64.7), and Saul finds that his appearance, too, has been changed (64.4, though this could, perhaps, be due to the disguise he has adopted, or to his being exceptionally haggard, rather than to a miracle).

The effect of this is to spread the occurrence of miracles rather more evenly than do the Bible, Josephus and especially Philo, in all of which there is a tendency for miracles to concentrate in the exodus—wilderness period. Whereas the exodus still remains paradigmatic for Pseudo-Philo, he appears to wish to stress that what God could do then he could continue to do subsequently, a point that is explicitly emphasized in the speeches attributed to Deborah (LAB 30, 32), and perhaps also Kenaz (25.6). The addition of explicit divine assistance for David (61.5–7) and involvement in summoning the shade of Samuel (64.7), carries the theme of the divine control of events right up to the end of Pseudo-Philo’s narrative.

The theme of divine control also emerges strongly when one examines to whom miracles are attributed in Pseudo-Philo. To be sure, we have already seen that both Josephus and Philo clearly regard miracles as acts of God, so that for them God is always the BNP. But if anything, Pseudo-Philo carries this even further so that the role of any human PNP or MNP almost vanishes altogether. In the majority of cases where Pseudo-Philo either narrates or has someone refer to a miracle, God is the only active miraculous agent mentioned (e.g. LAB 6.17; 8.1–3; 10.1, 7; 11.4–5; 15.6; 16.1–7; 20.4, 8; 23.7, 9–10; 25.6; 28.1; 30.5, 7; 31.2; 32.1, 5, 7–8, 13; 42.3; 44.9; 47.12; 51.6; 55.4–8; 64.7). In several other instances where angels act as BNP or MNP, one assumes that they do so as divine proxies (e.g. 26.8; 27.7–9; 35.6–7; 38.3–4; 42.9; 61.5–7, 9). Indeed, in several of these instances it is explicitly stated that God sent the angel in question.

There are not many cases where a human being acts as a genuine MNP, and even fewer where one acts as a PNP. Since the majority of those that do occur do so with Moses in the MNP/PNP role, it will be convenient to start with those, and then go on to the others.

The case of Moses is admittedly something of a puzzle, since although in summary statements Pseudo-Philo sometimes talks of miracles worked through Moses (e.g. LAB 9.7; 12.2) or his rod (LAB 19.11), Moses makes little explicit contribution when miracles are actually narrated. Thus, for example, he is given no role whatsoever in the plagues (10.1), the manna (10.7), or the punishment of Korah’s rebellion (16.1–7). In some other instances where Moses does have a role to play, it is not clear that it is strictly the role of mediator of numinous power. Thus, for example, Moses carries out his instructions to deposit Aaron’s rod in the sanctuary, but this does not really make him a mediator of the numinous power that causes the rod to blossom, thereby indicating God’s choice for the priesthood (17.1–4). Likewise, Moses is carrying the stone tablets when the writing disappears from them, but this is not to say that he has any part in his disappearance (12.5). After smashing the tablets and then the golden calf, Moses forces the people to drink the water mixed with golden calf dust, but it is not clear that he has any role in the outcome of this ordeal, in which the faces of the innocent shine, but the tongues of the willing idolaters are cut off (12.7). Moses perhaps acted as MNP by casting in the branch to sweeten the waters of Marah (11.15), but he is given no credit for the fact that this well subsequently follows the Israelites around in the wilderness, and elsewhere this is explicitly said to be the gift of God (10.7; 20.8).

The one place where Moses appears to act both as MNP and PNP in Pseudo-Philo is in the account of the Red Sea Crossing. When the children of Israel are hemmed in at the Red Sea, each group of four tribes proposes an inappropriate solution to their plight, and then Moses calls upon God (LAB 10.2–4). In response, God tells Moses to raise his rod and strike the sea with it. Moses does so, but the text goes on to emphasize that it is God who makes a path through the sea for the Israelites:

Moses did all this and God raged at the sea and it dried up. The streams of water stood up and the depths of the earth became visible, and the foundations of the world were laid bare by the fearful roar of God and by the breath of anger of the Lord (LAB 10.5).

The Israelites pass safely through the sea and the Egyptians pursue them, God befuddling the latter so that they would not realize where they were. Then,

While the Egyptians were in the sea, God again commanded the sea and said to Moses, ‘Strike the sea again’. And he did so. The Lord commanded the sea, and it returned to its course and covered the Egyptians and their chariots and horsemen to this day (LAB 10.6).

Insofar as Moses’ appeal to God prepares the BNP to act, he may rightly be judged to be a PNP here. It is less clear how far he is a genuine MNP, since although Moses strikes the sea with his rod on each occasion, on each occasion it is expressly God’s actions that cause the movements of the sea. It may that Pseudo-Philo includes Moses’ actions with the rod simply because they are part of the biblical narrative (Exod. 14:16–28). Although the biblical text makes no mention of Moses striking the sea with his rod, the tradition that he did so appears to have been widespread, and Pseudo-Philo may simply have followed it. Nothing at this point in the narrative indicates that either Moses or his rod acted as a mediator of God’s numinous power to effect the miracle. Since in each case Moses wields the rod at God’s command, it is hard to see how Moses’ striking the sea can be seen as prompting God’s action, as it might otherwise seem if one read the sequence of events described with the logic of post hoc ergo propter hoc. It is just conceivable that Pseudo-Philo thought God commanded the sea to retreat and then return when Moses struck it, but the text could equally well be read as suggesting that Moses’ actions with the rod were purely symbolic, perhaps to demonstrate his trust in God’s word both to God and to the watching Israelites, or to emphasize to the onlookers that the movements of the sea were God’s work and no mere freaks of nature.

If Moses’ actions with the rod seem almost otiose, this is curiously at odds with the statements that God will work or has worked great wonders through Moses (LAB 9.7; 12.2) and his rod (LAB 19.11). This may simply be an indication that Pseudo-Philo has worked with different traditions and has failed to combine them into a fully coherent account. Alternatively, it may result from the fact that Pseudo-Philo wishes to stress two different aspects that are in some tension with each other. On the one hand, when miracles are described, God’s role is emphasized at the expense of that of any human intermediary in order to stress the divine initiative and divine control of events.31 On the other hand, Pseudo-Philo is also concerned with good leadership. Pseudo-Philo concurs with Jewish tradition in general that Moses is the paradigmatic good leader of Israel. Since for Pseudo-Philo it is important to stress that Israel does better under good leaders, it is appropriate for him to stress that God worked special signs and wonders for Israel under the best of leaders. The summary statements thus stress the connexion between Moses and wonders, while the actual descriptions of miracles stress that it is God who is fully and uniquely in control.

The desire to associate God’s saving actions with good leadership in Israel may explain why Pseudo-Philo also attaches miracle stories to Kenaz and David, and emphasizes the miracle that took place for Deborah (see above), since these three are among the leaders he rates most highly. But when one examines the miracle stories narrated of these and other figures besides Moses, one again finds very few where a human being is given any significant role as an MNP or PNP.

At LAB 27.7 Kenaz does indeed pray for a wonder, and two verses later he not only draws his shining sword but is clothed with the spirit of the Lord, so here he does clearly act as both PNP and BNP (although he is assisted in the latter role by a pair of angels, which reduces the significance of his own contribution). When it subsequently requires the blood of an Amorite on Kenaz’s hand to release Kenaz’s sword from his grip (27.11), it is not clear in what sense this is strictly a miracle in which anyone acts as BNP, MNP or PNP.34 David perhaps acts as BNP in singing a song to ward off the evil spirit that afflicts Saul (LAB 60), though this is also less than clear. Eluma (LAB 42.2–3) and Hannah (50.4) both pray for children, and to that extent function as PNPs since their prayers are answered, but it would be stretching a point to say that they also act as MNPs on the grounds that God works through them to produce the children. The point is rather that God acts in each case to open a barren womb, so that it is once again the divine intervention that is stressed. It would also be stretching a point too far to call Gideon a PNP on account of his requesting a sign from the angel (LAB 35.6–7). Deborah and her fellow Israelites perhaps acted as joint PNP when God commanded the stars to smite Sisera’s army (LAB 32.11), though elsewhere it seems more plausible to see crying out to God as a general plea for help rather than specifically prevailing upon a BNP to act in a miraculous fashion. In each of the foregoing cases success as a PNP hardly makes the human petitioners into miracle-workers; it rather serves to emphasize their dependence on and trust in God.

The only exception that appears to give a human being a more positive role in miracle-working occurs in Deborah’s description of Joshua:

Joshua said to the sun and moon, ‘You, who have been made ministers between the Lord and his children, behold now, when the battle is still going on, are you abandoning your duties? Therefore stand still today and give light to his children and darkness to his enemies’. And they did so, as they had been commanded (LAB 32.10).

Taken in isolation, this passage appears to make Joshua an immanent, human BNP who can command the obedience of the sun and moon. This impression is lessened slightly when the passage is taken in conjunction with the previous verse, in which Deborah states that when Moses was dying, God made the contents of the firmament Israel’s servants; but this concept is odd, and it still leaves Joshua playing the part of MNP in a way that is unparalleled elsewhere in Pseudo-Philo. The problem is further compounded by the fact that previously Deborah attributes the standing still of the sun and moon solely to God: ‘and for your sake he [sc. God] commanded the luminaries, and they stood still in their own places; and when your enemies came against you, he rained down hailstones on them and destroyed them’ (LAB 30.5). In the biblical account Joshua appears to address both the heavenly bodies and God (Josh. 10:12), though the Bible appears to regard Joshua as acting as PNP here (Josh. 10:14, ‘There has been no day like it before or since, when the LORD hearkened to the voice of a man; for the LORD fought for Israel’). One might say that at LAB 32.10 Pseudo-Philo more closely reflects the biblical account insofar as he has Joshua address the sun and moon, whereas at 30.5 he more closely reflects his own tendency to confine all miracle-working to God. Or again, one may appeal to the possibility that Pseudo-Philo wishes both to emphasize divine control of events and to associate a beneficent miracle (from the Israelite perspective) with a good leader. But it must be admitted that there is no solution to the difficulty here that is entirely satisfactory.

This one exception apart, the tendency in Pseudo-Philo is clear enough. Miracles are acts of God in which humans have little role to play, except insofar as they act in obedience to divine commands or express radical trust in God through their petitions. Even then, it is God who performs the effective action. On the other hand, these miraculous interventions are not confined to any one period in Israel’s past, such as the exodus, however paradigmatic such a period may be. God intervenes for his covenant people wherever there is a good leader to provide and promote an acceptable level of obedience. This leaves open the possibility that God may so act again in the future, provided another such good leader arises.

The Function of the Miraculous in Pseudo-Philo

Many of the functions of the miraculous in Pseudo-Philo have already emerged in the course of discussion. Pseudo-Philo is concerned to show that God is firmly in control of history. Miracles are acts of power that show God to be firmly in control. With few exceptions, miracles are performed on divine initiative with very little effective contribution from humans. They may vary in scale or impressiveness, but there appears to be no restriction on what God may do. If a miracle is needed in order for God to achieve his purpose, then a miracle is worked. Miracles are then but the most spectacular way in which God intervenes to control events.

This can be made more specific, since for Pseudo-Philo, God has three main aims: to punish sinners, to reward or protect the righteous, and to stand by the covenant he made with the patriarchs and thus to preserve Israel however much Israel sins.39 Pseudo-Philo’s notion of God’s retributive justice is even more thoroughgoing than that of Josephus. Whereas in general terms Josephus wants to show that those who follow God’s commands prosper, while those who rebel suffer, Pseudo-Philo is loath to allow any exception to this Deuteronomistic rule. Accordingly, where the biblical story appears to present an innocent sufferer, such as the Levite’s concubine who dies following an all-night gang rape (Judg. 19:1–28), Pseudo-Philo has to attribute a previous dire sin to her so that she only got what she deserved (LAB 45.3). Where there is an apparent exception to this, such as Jephthah’s daughter Seila who must be sacrificed on account of Jephthah’s foolish vow, Pseudo-Philo assures his reader that she will be justly recompensed in the after-life (LAB 40.4), just as Gideon will receive post-mortem punishment for his idolatry, which God cannot punish in this life (LAB 36.4).

One function of the miraculous in Pseudo-Philo is thus to ensure that the righteous are rescued from peril and the wicked punished. Pseudo-Philo is often also concerned that the punishment should fit the crime. Thus, for example, not only are Abraham and the faithful non-idolaters at the time of Jair miraculously rescued from the flames employed by those who would punish them, but their rescuers miraculously perish by fire (LAB 6.15–18; 38.3–4). Micah’s mother, who persuaded her son to set up as an idolater to make a name for himself, is punished by rotting and worms while her son is consigned to fire (LAB 44.9; 47.12); not only do they then take to blaming each other, but Micah’s idols are to become the instruments of their torment. The willing idolaters who misused their tongues to persuade Aaron to make the golden calf have their tongues severed by the drinking ordeal (LAB 12.7).

Miracles are also often worked for the preservation of the covenant people. Several times the miracle is simply the opening of a previously barren womb in order to ensure that the line continues or that a much-needed leader is born (LAB 8.1–3; 32.5; 42.3; 50.4–51.1). Even more often God lends miraculous aid to secure victory for Israel in battle, or liberation from her oppressors (LAB 10.1, 5–6; 15.6; 27.7–9; 30.5; 31.2; 32.10, 11; 61.5–8). The other type of salvific miracle, the miracle of provision, occurs in Pseudo-Philo only in connexion with the wilderness wanderings, namely the manna (and quails) and the well (LAB 10.7; 11.15). Unlike the biblical account, and unlike Philo and Josephus, Pseudo-Philo envisages the well following the Israelites on their wanderings (cf. 1 Cor. 10:4–5). In common with most Jewish references to the manna story, but in distinction from the Gospel feeding stories, Pseudo-Philo emphasizes that the manna is something that came down from heaven (LAB 10.7; 19.10). Similarly the crossing of the Red Sea in Pseudo-Philo (LAB 10.5–6), again in common with most other Jewish accounts of this event, emphasizes elements that are lacking in the Gospel sea-crossing miracles, namely the parting of the waters, the passage through on dry land, the drowning of the Egyptians, and the role of Moses’ rod.

The third main class of miracle in Pseudo-Philo, after the punishment and salvation miracle, is the sign-miracle. This has already been discussed above under Pseudo-Philo’s use of the term signum in connexion with the miraculous signs given to Gideon (LAB 35.6) and Kenaz (LAB 27.7). A few other miracles function as signs in Pseudo-Philo even when the word signum is not present. The blossoming of Aaron’s rod to designate the priestly family is perhaps the clearest example (LAB 17.1–4). The ignition of Manoah’s sacrifice by the angel may be another (LAB 42.9; cf. Judg. 13:15–21), especially since Manoah has previously complained, ‘Behold, am I not worthy to hear the signs and wonders that God has done among us or to see the face of his messenger?’ (LAB 42.5). The return of the ark to Judaea after the Philistines set the cart carrying it at a cross-roads (LAB 55.6–8) is another sign, although the repeated prostration of Dagon before the ark and the illnesses accompanying its presence in the Philistine temple (LAB 55.3–6) combine punishment with signs of divine displeasure. The same perhaps may be said of the disappearance of the writing from the tablets of the law when Moses descends the mountain only to encounter the golden calf (LAB 12.5). The main function of these signs in Pseudo-Philo is to demonstrate divine control of events through providing the means by which good characters may know that they are acting within the divine plan.

Unlike Philo and Josephus, Pseudo-Philo envisages a class of eschatological miracles (LAB 27.7; 28.1). These are apparently revealed to Kenaz, but it is not specified precisely what they are. At LAB 27.7 Kenaz prays, ‘Lord God of our fathers, you have shown your servant the wonders that you are ready to do for your people in the last days.’ He goes on to pray for a wonder that will help him defeat the enemy. At LAB 28.1 Kenaz informs his audience, ‘Behold now the Lord has shown me all his wonders that he has prepared to do for his people in the last days.’ The phrase ‘wonders for his people’ strongly suggests the type of liberating, military intervention for the defeat of Israel’s enemies and oppressors worked at the time of the exodus (cf. LAB 9.7; 12.2; 20.4; 30.7), and this is further borne out by the type of wonder Kenaz immediately goes on to request at 27.7. It may be, then, that Pseudo-Philo is similar to Wisdom in this regard. On the other hand Pseudo-Philo may have in mind the end-time expectations associated with the precious stones at LAB 26.13. This passage predicts both the loss of the stones along with the destruction of the temple, and their ultimate restoration along with ‘many others, much better ones’ when God comes to visit the earth (presumably in judgment). Then the light of these stones will replace that of the sun and moon for the just. These two possibilities are not incompatible, since God’s visitation of the earth would presumably include the final defeat of his (and Israel’s) enemies.

It is interesting that these predictions are associated with the figure of Kenaz. Whereas in the Hebrew Bible the yearning seems to be for a king like David, in Pseudo-Philo the people want someone ‘who can rule us as Cenaz did’ (LAB 49.1). Pseudo-Philo’s Kenaz is a descendent of Judah (according to LAB 25.2 he is of the family of Caleb; according to 15.3 Caleb is descended from Judah). The book ends with another descendent of Judah, namely David, about to ascend the throne, having been anointed for the task by Samuel (LAB 59.1–4), who was in turn the immediate answer to the people’s pray for a ruler like Kenaz. This at least raises the question whether Kenaz might function as a type for the Messiah in Pseudo-Philo, in which case we should have important but rare evidence for the association of Messiah and miracles in Judaism. The scholarly consensus appears to be against seeing much messianism in Pseudo-Philo, however. Thus, for example, Murphy states, ‘The strongest argument against the idea that Pseudo-Philo expects an eschatological Messiah is that when the text speaks of the eschaton, no Messiah is present.’ Charles Perrot is equally emphatic that ‘Le Pseudo-Philon ne fait aucune allusion à un messie humain’.48 Jacobson is only slightly less categorical on the issue: while holding that ‘there is little evidence that he [sc. Pseudo-Philo] thought in terms of a Messiah’, he does review what little evidence there is and comes to the conclusion that ‘it seems that LAB may have had a notion of a messianic descendant of David, but if he gave the concept any serious thought it is not evident. The notion in LAB, if such there is, is quite bare and not fleshed out at all’.

One could think of at least two good (and mutually compatible) reasons why Pseudo-Philo might refer to a Messiah in somewhat vague terms: either the idea may be sufficiently familiar to his target audience that it does not need spelling out (compare Pseudo-Philo’s treatment of parts of the biblical narrative), or else Pseudo-Philo’s political situation is such that he needs to be circumspect about what he says and content himself with mere hints. As Pseudo-Philo’s narrative now ends, the reader is left expecting the reign of David, the Lord’s anointed (LAB 51.6; 57.4). But the fact remains that nowhere does Pseudo-Philo talk of a Messiah as God’s eschatological agent. The expectation created by the ending of LAB is thus not so much of an eschatological Messiah, but of another good leader like David, Kenaz or Moses: perhaps ‘a messiah’ in the sense of one commissioned by God to lead his people in time of crisis, but not ‘the Messiah’ in any eschatological sense. Such a reading (a messiah rather than the Messiah) fits better with Pseudo-Philo’s pattern of a succession of (good and bad) leaders within God’s continuing control of history.

Yet the possibility of a messianic interpretation of Pseudo-Philo would be especially intriguing in connexion with the exorcism that David performs for Saul in LAB 60. Of immediate interest is the presence in David’s song of exorcism of the threat to the evil spirit, ‘But the new womb, from which I was born, will rebuke you, from which in time one will be born from my loins and will rule over you’ (LAB 60.3). It is perfectly possible to see this as a reference to David’s son Solomon, who elsewhere seems to have been credited with magical powers and control over demons. But the reference to one descended from David could also be understood messianically. In that case we should have a virtually unique allusion in Jewish literature of the period to the idea that the Messiah would have control over evil spirits, which would obviously be of considerable significance in an evaluation of Jesus’ exorcistic activity. Unfortunately, the very uniqueness of this allusion makes the Solomonic interpretation appear more likely.52

Even it LAB 60 cannot be interpreted messianically, it does provide a loose link with the Gospel traditions about Jesus simply by virtue of being an exorcism story. Pseudo-Philo shows virtually no interest in healing or exorcism miracles elsewhere (although, as with Philo and Josephus, there are not that many in the biblical story he is retelling). The only other place where healing or exorcism is mentioned in Pseudo-Philo is in the description of the precious stones discovered by the sinners from the tribe of Asher:

These are the precious stones that the Amorites had in their sanctuary … Among them the brightest was the one that was cut in the fashion of open-work and purified people of demons. For even if one of the Amorites was blind, he would go and put his eyes on it and regain his sight (LAB 25.12).

This is noteworthy for its apparent connexion of demonic attack with blindness, but it must be regarded as magical rather than miraculous. There is no indication that God is the BNP curing blindness and expelling demons through these stones; on the contrary, they have been used in idolatrous worship (LAB 25.11; 26.4). Moreover, they have to be removed especially by God since Kenaz does not have the means to destroy them, but they are too dangerous to keep (LAB 26.4–8). These stones thus represent forbidden powers, not legitimate miracles.

Pseudo-Philo also displays a further (negative) interest in magic in the incident with Aod, who misleads the Israelites by showing them the sun at night, and claiming it to be a demonstration of the power of the Midianite gods (LAB 34). Pseudo-Philo states several times that the Midianite priest Aod in fact works through magic (magica/maleficia), and, more specifically, through the offices of rebel angels who have lost their share in the world to come (LAB 34.2–3). This may well be an echo of the traditions found in 1 Enoch (see, e.g., 1 En. 6–9), while the figure of Aod may have been created by Pseudo-Philo from the warning against the false prophet who may lead people astray with signs and wonders at Deut. 13:1–3, together with the notice at Judg. 6:1 that ‘The people of Israel did what was evil in the sight of the LORD; and the LORD gave them into the hand of Midian seven years.’ Deuteronomy 13:3 suggests that if such a false prophet should arise, it is because ‘the LORD your God is testing you, to know whether you love the LORD your God with all your heart and with all your soul’. Similarly, Pseudo-Philo states in connexion with Aod’s deception that ‘God wished to test if Israel was still in its sinfulness’ (LAB 34.5). In any case, the notion that Aod can carry out his deception only because God permits it as a test is fully consistent with Pseudo-Philo’s insistence on the divine control of events.

The same is just as true of the final piece of magic that occurs in Pseudo-Philo. At LAB 64 Saul goes to the Witch of Endor in order that she may conjure up the shade of Samuel for him to consult. She appears to succeed in her necromancy, but when Samuel speaks he tells both her and the doomed king that it is not they who have raised him up from the dead but rather God’s command (64.7). This piece of magic thus turns out to be a miracle after all.

The final question is how far Pseudo-Philo actually believes in the miracle stories he narrates. Unfortunately, an answer is not so easy to come by as in the case of Josephus and Philo, since, unlike them, Pseudo-Philo makes no explicit reference to the question. On the one hand one could argue that his insistence on divine control of events, of which miracles constitute the most spectacular examples, would be largely hollow unless he took the miracles seriously as real events. On the other hand one could equally well argue that the liberties Pseudo-Philo takes with the biblical narrative indicate that he is not all that concerned with literal historicity. To be sure, Pseudo-Philo may well be drawing on older haggadic traditions rather than simply inventing variations from the biblical narrative himself, but it must be said that he is less faithful to the biblical text than either Josephus or Philo, who also draw on other traditions. In the case of the precious stones uncovered by Kenaz, and the ones subsequently sent to replace them, their function of giving light in connexion with the cult may have more to do with the illumination supplied to Israel by the torah than with any literal miracle. One might suppose that a first-century ce Jew would be likely to take the biblical narrative seriously, including its miracle stories, but this would be to argue in a circle if the purpose is to discover the attitude of first-century ce Jews towards miracle. The purpose of Pseudo-Philo’s LAB seems not to be to inform its audience of facts they did not know before, but to interpret well-known traditions in a way that would encourage its audience in a discouraging situation. Given this rhetorical purpose and the absence of any explicit discussion of the credibility of miracles in LAB, what the real author of LAB thought about the question is an issue his text simply does not answer.


In common with Josephus and Philo, Pseudo-Philo shows very little interest in the types of miracle that predominate in the Gospel accounts of Jesus, namely healing and exorcism. This must be at least partly the result of retelling a section of the biblical narrative in which few such miracles occur. As, for example, the lengthy section on Kenaz shows, Pseudo-Philo is prepared to add substantial extra-biblical material to his account, and this could have afforded him some scope for inserting healing miracles if these interested him. It is not hard to see why they might not have done however: Pseudo-Philo’s concern to present a strictly Deuteronomistic view of rewards and punishments would have encouraged him to view illness as a punishment for sin (as he does with Micah’s mother); there would then be something a little odd in presenting God either directly or indirectly healing an illness that God himself had sent as a punishment (although, as we have seen, Wisdom seems capable of such an oddity). In any case Pseudo-Philo’s presentation of biblical history makes him primarily concerned with God’s intervention against human wickedness and oppression rather than with disease.

Pseudo-Philo’s universe does contain rebel angels and evil spirits, but he does not show a great deal of interest in them, and so gives them little scope to act as enemies of God’s people. There is thus little scope for any miraculous intervention against demonic oppression. Nevertheless, in the one place where demonic attack and apotropaic countermeasures are suggested by the biblical text (1 Sam. 16:14–23), Pseudo-Philo does develop it by providing the details of David’s exorcistic psalm. This psalm contains what could conceivably be a hint that Pseudo-Philo saw power over demons as a messianic function, but it is much more likely to be a reference to Solomon’s exorcistic abilities. Otherwise, both the miracles Pseudo-Philo narrates and those he appears to expect at the end time are mainly for the punishment of sinners and the salvation, either of righteous individuals, or the covenant people from their human oppressors. For Pseudo-Philo, God’s miraculous interventions against Israel’s enemies in the periods of the exodus and the judges are paradigmatic for the interventions hoped for in the future. Nonetheless, the very presence of such eschatological hopes, together with the Davidic exorcism, make Pseudo-Philo’s views on miracle perhaps a little closer to the Gospels than either Philo or Josephus.

Compared with the Greek-language literature of Josephus, Philo and the Wisdom of Solomon, Pseudo-Philo and Ben Sira exhibit far less awareness of ‘miracle’ as a distinct category of event that needs to be explained or justified. This is hardly surprising given that the three Greek-language authors were in one way or another far more concerned to justify Jewish traditions in a Hellenistic environment than were the two Hebrew-language authors so far considered. This distinction between ‘Hellenistic’ and ‘Palestinian’ Judaism should not be pressed too far, however. On the one hand Josephus can shade the miraculous into the merely providential, while on the other Ben Sira can celebrate the signs and wonders of Elijah and Elisha as their most distinctive deeds. Again, there is a certain family resemblance between Josephus’s theory of divine providence as retributive justice and the function of divine interventions to reward and punish in both the Wisdom of Solomon and Pseudo-Philo (a family resemblance that no doubt derives from a common ancestor in Deuteronomistic theology).

Pseudo-Philo and Ben Sira provide two Palestinian Jewish perspectives on the question of miracle that may be fairly representative of the ‘main-stream’ Judaism of their time (assuming there was such a thing). A further substantial Palestinian Jewish text worth looking at (though more for its demonology that its miracles) is the book of Jubilees, but this appears to belong in a stream related to both Enochic and Qumran literature. It is in this context that the next chapter will go on to examine Jubilees, having first reviewed the relevant mythological background in 1 Enoch.

Chapter 6

Enochic Literature


After Josephus, Philo and Pseudo-Philo, the other most substantial surviving text that retells some part of the biblical narrative is the book of Jubilees. It would thus seem appropriate to tackle this next. At this point, however, two factors indicate a change in the manner of treatment. The first is that Jubilees shows insufficient interest in miracle to make it either worthwhile or even feasible to treat it in the same manner as the previous texts. The second is that it is closely related both to 1 Enoch and to the Genesis Apocryphon and other Qumran literature. Although Jubilees does not itself appear to be a sectarian text, it was used by the Qumran community (fragments of Jubilees have been found in Qumran caves, and the work is cited by cd [the Damascus Document]) and shares a number of ideas with the Qumran sectarian texts, not least insistence on a solar calendar with a 364-day year. 1 Enoch also espouses this calendar, and, as we shall see, many of its ideas on demonology are taken up in Jubilees.

Gabriele Boccaccini has put forward the interesting thesis that there was a distinctive current of Enochic Judaism, characterized by a particular view of the origin of evil and a reserved attitude to the Second Temple, and that this Enochic Judaism was in fact one and the same as the Jewish party referred to as the Essenes by ancient writers such as Philo, Josephus and Pliny. This hypothesis is both attractive and plausible, and whether or not one accepts it in its entirely it provides a useful model for situating a number of texts that do seem to be related. On this model the books collected in 1 Enoch represent some of the literature of the mainstream Enochic (or Essene) party. Jubilees also belongs to this group, but in the wake of the Maccabaean crisis it has absorbed some important ideas (such as the importance of the Mosaic torah and Jewish covenant particularism) from the opposing (Zadokite) party. The Qumran sect represents a fringe extreme that split off from Enochic Judaism after the writing of Jubilees but before the Book of Similitudes took shape (hence the absence of the latter from the Aramaic fragments of 1 Enoch recovered at Qumran). The Qumran library comprised both sectarian literature and some of the works (such as Jubilees and the earlier parts of 1 Enoch) that were valued in its parent body prior to the Qumranic schism. After the schism, there ceased to be any exchange of literature between Qumran and the wider Enochic/ Essene party.

This chapter and the following one will adopt this model to the extent of focusing on what ‘Enochic’ literature may have to contribute both to Jewish views on miracle in general, and in particular to notions of demonic origins that may form an important part of the context of Jesus’ exorcisms. The present chapter will examine the Book of Watchers (1 En. 1–36) and the book of Jubilees. The next will turn to the Qumran material. Both chapters will be dealing with texts that are rather different from what has gone before.

The Book of Watchers in 1 Enoch

1 Enoch is not the most obvious text to search for Jewish views on miracle, since it contains virtually nothing that would count as a miracle in the relevant sense. Indeed, the ‘nature sermon’ in 1 Enoch 2–5 suggests that the Judaism of 1 Enoch saw God’s sovereignty more in the regular workings of nature than in divinely caused exceptions to that order, for the sermon contrasts the disobedience of humans with the obedience of the sky, weather and plants to the divine laws. The quasi-scientific interests of the Astronomical Book (1 En. 72–82) and the concern with the regular 364-day calendar could well point in the same direction. Although 1 Enoch is a composite text, a collection of books composed over the course of two centuries or more, and although it encompasses heaven, earth and the abyss, God, angels and human beings, protology, eschatology, and much of history in between, it consistently lacks anything resembling a miracle story or a discussion of the miraculous.

1 Enoch nevertheless expresses a distinctive view of the origins of evil, and of evil spirits, that provides a possible context for understanding the practice of exorcism. This is not only of some interest in itself but also of considerable importance in discussing other texts that have been influenced by similar views.

The work now known as First or Ethiopic Enoch has been preserved in its entirety only in Ethiopic manuscripts, although some parts of it have been found in Greek translation, a small section in Latin, and in Aramaic fragments at Qumran. It is, as stated above, a composite text, comprising five books: the Book of Watchers (1 En. 1–36), the Book of Similitudes (1 En. 37–71), the Astronomical Book (1 En. 72–82), the Book of Dreams (1 En. 83–90), and the Epistle of Enoch (1 En. 91–107), of which the Book of Watchers and the Astronomical Book appear to be the oldest sections, dating from perhaps the third century bce, while the Book of Similitudes is probably the latest, dating perhaps from the first century ce. The Aramaic fragments discovered at Qumran contain parts of four of these sections, but not the Book of Similitudes. The Qumran fragments also contain remnants of what appears to be a work related to the Manichaean Book of Giants, which thus appears also to have been part of the Enochic corpus. The Aramaic fragments of the Book of Watchers indicate that the version used at Qumran has substantially the same form as the later Ethiopic text.

This Book of Watchers contains what proved to be an influential myth of the origin of evil, based on the cryptic account of the ‘sons of God’ and the Nephilim and ‘mighty men’ in Gen. 6:1–4. In the Book of Watchers, the angels, or ‘Watchers’, mate with human females to produce offspring that are notable mainly for their rapacity and violence. God subsequently acts to punish both the Watchers and their giant offspring. The giants’ bodies are destroyed but their spirits live on—as the evil spirits or demons. The fact that the giants have a half-human ancestry and that their spirits were formerly embodied may explain why the postdiluvian evil spirits seek to re-enter human bodies and possess them. As we shall go on to see in the next section, a very similar account of this myth is given in Jubilees, and it is referred to or presupposed in a number of other texts.

Assessment of this myth is complicated by the fact that the Book of Watchers is itself composite. The main account of the fall and punishment of the Watchers is given in chapters 6–11, in which Enoch is not mentioned. 1 Enoch 12–16 contains an account of Enoch’s being asked to intercede on behalf of the Watchers and of the answer he is commanded to give them. This presupposes the narrative of 6–11 but fits awkwardly with it, since Enoch seems partly to be duplicating a function already performed by (non-rebellious) angels, namely that of announcing God’s verdict on the rebellious Watchers. A further discontinuity is that 1 Enoch 6–11 simply narrates events, whereas most of the events of chapters 12–16 are contained within Enoch’s vision. Chapters 17–36 then contain a loose continuation of Enoch’s visionary experiences in the form of a description of his tour of the Earth and the places of punishment. The whole collection is preceded by an introduction, chapters 1–5, in which Enoch warns of the coming judgment (1) and preaches a brief sermon contrasting the obedience of nature with the disobedience of humanity (2–5).

Even within chapters 6–11 there appears to be a fusing of different traditions (though not necessarily a combination of previously extant texts). The story in chapters 6 and 7 concerns two hundred rebel angels, led by Semhazah, who bind themselves by an oath to take wives from among the daughters of human beings. The offspring of this illicit union are the giants, who set about devouring first the vegetable produce of the earth and then animal life, including human beings. Chapter 8 introduces a different angelic leader, Asael, with another band of angels whose sin is to reveal forbidden knowledge to human beings in the form of weaponry, cosmetics, astrology and divination (presumably thereby encouraging the human sins of violence, lust, magic and idolatry). In chapter 9 a trio of faithful angels report the misdeeds of both Asael and Semhazah to God, and chapter 10 proceeds to describe their punishment. Raphael is sent to bind Asael and bury him in a hole in the desert. Gabriel, meanwhile, is despatched to foment mutual destruction among the giants so that they will perish at one another’s hands, while Michael is instructed to tell the Watchers what their children’s fate will be and then to bind them (the Watchers) under the Earth until the day of judgment. As a further complication the angels led by Semhazah are also guilty of revealing forbidden knowledge of magical medicine to their wives.

At the beginning of chapter 10 God sends the first of the four angels, Asuryal, to the son of Lamech (i.e. Noah), to warn him of the coming flood, so that Noah and his seed might be preserved from the coming destruction. It is not immediately apparent, however, who or what the flood is intended to destroy. It can hardly be the Watchers, since these are to be imprisoned until the last judgment. Neither is a flood needed to destroy the giants, since they are already bringing about their own mutual destruction (1 En. 10.9; 14.6). It would be odd if God were sending a flood to destroy the humans and animals who were the helpless victims of the giants’ depredations, so perhaps the purpose of the flood is to destroy a world that has been thoroughly corrupted by Asael’s teachings (10.8).

In any case neither the Watchers nor their giant offspring are available to play any demonic role from that point on, since the former have all been imprisoned and the latter slain. Yet evil persists; the apparent explanation for this is given in the course of Enoch’s vision at 15.8–16.1. The giants become evil spirits (15.8), or more precisely, evil spirits proceed from their bodies (15.9). These evil spirits will remain upon the earth, oppressing humanity and corrupting the earth until the day of judgment (15.9–16.1); the mode of their final demise is not stated. The precise nature of this transmogrification from giants into evil spirits is also unclear. Presumably one is meant to suppose that the evil spirits come forth from the giants after they have slain one another by the sword. It may be that this stubbornly surviving element is due to the giants’ angelic paternity, although the notion that demons are the souls of ordinary dead human beings, especially wicked ones, is also attested (e.g. by Josephus at War 7.185, see Chapter 12 Section 3.a below), so that angelic paternity would not necessarily have been seen as essential to the giants’ extra-corporeal survival.

The persistence of the evil spirits at 15.8–16.1 is in any case in some tension with the rosy prospects for a renewed, cleansed earth described at 10.17–11.2. To be sure, this must be read as a description of eschatological renewal after the last judgment (10.12–16) and the text simply passes over in silence anything that occurs between the binding of the Watchers and their final punishments (beyond describing it is as a period of 70 generations, 10.12), but there is no indication in this version of the story that the giants survive their mutual destruction in any form. Moreover, although it could be argued that the passage describing the generation of evil spirits (1 En. 15.8–16.1) fits its context well enough as an account of the consequences of the angels’ illicit union with human women, which has just been condemned (15.3–7), the text would also make reasonable sense without it. There is thus some plausibility in the suggestion that the introduction of evil spirits at this point represents a scribal addition to the fallen angel tradition, designed either to incorporate popular ideas about evil spirits, or to systematize the origin of various types of evil, or (perhaps most likely) both. The Watchers are thus shown to be the originators of both human and inhuman evil. Through their teaching they introduce humankind to war, sexual temptation and magic;18 through their progeny they give rise to the evil spirits responsible for disease and other types of human ill beyond human control.

The Qumran Book of Giants is a further witness to this aetiology of demons. Particularly noteworthy among the fragments of this text is 4Q531 fragment 14, which indicates that the giants will not survive in their present form as bones or flesh (perhaps implying that after the flood they will be able to exist only as evil spirits). The text 4Q530 2.1–3 perhaps also points to the survival of the giants’ souls. The surviving fragments of the Book of Giants do not appear to envisage the giants killing one another, but may indicate that they have dream visions of perishing in the flood (4Q530 2.7–12; 6Q8 2).

This myth of demonic origins hardly took place in a social and political vacuum. It has been argued, for example, that the story of illicit sexual relations between angels and human women represents concern about Jewish priests taking unsuitable marriage partners. It has also been suggested that the bellicose behaviour of the giants bears a pointed resemblance to that of Alexander’s successors, the Diadochoi, with the angelic paternity of the giants perhaps mocking the Diadochoi’s claim to divine paternity.24 This would not mean that the myth is merely a political comment in allegorical guise (like an ancient equivalent of George Orwell’s Animal Farm). Neither should one assume that the contemporary concerns of the people who contributed to various stages of the formation of the Book of Watchers determine the meaning of the text for all subsequent readers. It may indicate, however, that the evils that most exercised people in the circle in which the Book of Watchers took shape were those of a social or political nature. Even if for these people evil had an irreducibly spiritual dimension in which rebellious angelic powers were somehow responsible for disorders on earth, the primary empirical manifestation of that evil seems to have been in human misbehaviour such as the violence of pagan kings or the impurity of priestly marriage arrangements. On the face of it this is a view of the demonic that is quite distinct from the (perhaps more popular) view that blames evil spirits for possession, illness and a number of other misfortunes that seem totally outside human control. As a first approximation it may be seen as analogous to the modern distinction between ‘moral evil’ and ‘natural evil’. Again, as a rough approximation, one may wish to distinguish between the demonic conceived as a malignant social force corrupting society through widespread human wrongdoing, and individual evil spirits capriciously attacking individual human victims, although these two conceptions should perhaps be seen as ideal types rather than descriptions of what distinct sets of people believed. One might perhaps characterize these ideal types as cosmic-social-moral versus earthly-individual-affliction.27 The Qumran conception of (good and) evil spirits influencing individual human behaviour clearly blurs this rough distinction in one way (since it concerns moral evil in individuals), and I shall shortly be drawing attention to another kind of blurring. But first there are two other points to be noted.

The first is that in the Book of Watchers the only solution to the evils described is presented in terms of a final judgment and cleansing of the earth. No hope is expressed for any repeat of miracles of liberation such as took place in the exodus tradition. But (as will be argued in Chapter 9) it is precisely such miracles of liberation that seemed to have excited the interest of those Jews who were most interested in miracles. Thus the absence of exodus references in 1 Enoch and the comparative lack of interest in miracles in the Enochic tradition may well be related.

The second is that it is the angels, not the evil spirits, who are given leading roles in the cosmic drama of the Book of Watchers. The evil spirits’ demise at the eschaton is presupposed but no emphasis is placed upon it; it is on the final demise of the Watchers that the emphasis falls. The binding and ultimate defeat of the Watchers is thus not equivalent to a binding and ultimate defeat of evil spirits, and the evil spirits play only a relatively minor role. Thus, not only does the Book of Watchers, and indeed 1 Enoch as a whole, contain no real exorcisms, it would also be a mistake to read a significant eschatological defeat of evil spirits directly from the text of the Book of Watchers.

Yet there is some potential significance here. The very fact that an account of the origin of evil spirits has been incorporated into an account of the fall of the Watchers and the evil deeds of their gigantic progeny creates a link between the two. That link could allow subsequent blurring between the two conceptions of the demonic outlined above, and indeed, as we shall see, in the book of Jubilees that blurring already starts to occur (and the role of the evil spirits or demons becomes more important). Thus although the Book of Watchers says nothing about exorcism, and does not itself contain the view that inflicting defeat on individual demons through exorcism would in any way contribute to the final overthrow of the evil cosmic powers, it does provide a possible starting point for such ideas to develop.

In relation to the Jewish context of Jesus’ miracles, the importance of the Book of Watchers (and the rest of 1 Enoch) probably lies more in its subsequent influence than in its immediate content. Whether or not Enochic ideas directly influenced the historical Jesus, it can hardly be doubted that they were to prove highly influential in the Christian tradition. Not only has 1 Enoch been preserved in its entirely only by the Coptic Church, but it is directly quoted in the New Testament at Jude 14–15 (citing 1 En. 1.9). The myth of the fallen Watchers is also presupposed at 1 Pet. 3:19–20; the fact that the reference there is so allusive suggests that the (probably Roman) author could simply assume that his Christian readers in Asia Minor would be familiar with the story of the Watchers. Moreover, 2 Pet. 2:2–5 also appears to refer to the story of the rebel angels punished at the time of the flood.

The influence of 1 Enoch on early Christian literature is traced in some detail by James VanderKam, who concludes:

The greatest contribution of the Enochic apocalyptic tradition to early Christian thought was its angelic reading of Gen 6:1–4. Its influence can be traced in the various centers of Christianity from New Testament times until the early fourth century. In one form or another it gave Christian writers ammunition for explaining the presence of evil, idolatry, and demons in the world and the certainty with which wickedness would be punished at the judgment.

That the Enochic traditions have made such a contribution to early Christian thought clearly makes them part of the relevant Jewish context for interpreting early Christian documents, if not also the historical Jesus. The immediate concern, however, is with the further development and use of the Enochic traditions within Judaism, and with that aim (at least partially) in mind we shall turn next to the book of Jubilees.


The book of Jubilees survives in complete form only in a number of Ethiopic texts, although about a quarter of it also survives in a Latin translation. Some Greek fragments survive in quotations by ancient authors, as do some fragments in Syriac. A number of Hebrew fragments of the text have been discovered at Qumran. It is generally accepted that the Ethiopic version is a translation from a Greek translation of a Hebrew original. But a comparison of the Ethiopic manuscripts with the Hebrew fragments from Qumran suggests that the Ethiopic is nonetheless a faithful reflection of that Hebrew original.32

The work almost certainly dates to the second century bce, either to the time of the Hellenizing crisis under the Seleucid ruler Antiochus IV (Epiphanes), or to the era of the Maccabaean successes that followed it, that is some time in the period 170–140 bce. The author was almost certainly a Palestinian Jew. Although Jubilees shares many points of agreement with the sectarian literature of Qumran, it does itself not appear to be a sectarian document, since it lacks any evidence of a breach with the Temple authorities or of a separation from other Jews. The most plausible explanation is that its author belonged to the Hasidim who were opposed to Hellenization, but wrote Jubilees before some of the Hasidim split off into the Qumran sect.

Jubilees takes the form of a revelation by the Angel of Presence to Moses on Mount Sinai. The content of this revelation is mainly a retelling of the history of the covenant people down to this point, and it thus essentially covers the same ground as the biblical Genesis and Exodus 1–14. One of its principal concerns is to exhibit the patriarchs, especially (a whitewashed) Jacob as models of behaviour and loyal observers of the torah (and hence to legitimate that torah). Jubilees accordingly introduces a number of halakoth into its version of the patriarchal narratives. The book also inveighs against any intercourse with Gentiles, and strongly promotes its version of the festal calendar. Calendrical concerns are also shown in the way Jubilees structures time into weeks of years and jubilees (periods of seven weeks of years), hence the title of the book.

It is perhaps not surprising that a work so concerned to promote an idealized revealed divine order and strict observance to the torah should have relatively little interest in the miraculous. To be sure, Jubilees’ God is very much in control, but he exercises that control more by revealing ordinances and demanding obedience than by intervening in events. It is the task of the higher ranks of angels to reveal those ordinances to humankind (or at least that section of it within the covenant) and of the lower angels to maintain the regular working of nature (Jub. 2.2). A desire for order is further indicated by the divinely revealed solar calendar in which each year contains exactly 52 weeks, each month 30 days, and each season 13 weeks plus one day. In this calendar the festivals fall on the same day of the week each year, a kind of monotonous regularity that perhaps aims to mimic an eternal order. The author’s polemic against mixing with Gentiles and insistence on maintaining ancestral customs, especially ritual ones, is perfectly understandable against the background of the attempted Hellenization of the Antiochan period. But it can also be understood, not simply as a desire to maintain racial and cultural identity in the face of a threat of assimilation, but as an attempt to maintain order in the face of the chaotic forces represented by the Gentiles. This is further suggested by the way in which the text tends to associate Gentiles with demonic forces (e.g. Jub. 11.4–6; 12.20; 15.31; 22.17). Although a miraculous intervention can act to restore a disrupted order (e.g. perhaps Jesus’ stilling of the storm in Mk 4:35–41 and parallels), most miracles can also be understood as interruptions of order. This may explain why Jubilees shows little enthusiasm for them.

That is not to say that Jubilees recounts no miracles at all. The main miracles of the exodus narrative are narrated, albeit in a highly compressed form. The ten plagues are described in a paragraph of four verses (Jub. 48.5–8), while the crossing of the Red Sea is reduced to a mere two (Jub. 48.13–14). There is also an allusion to the competition between Moses and the Egyptian magicians to convince Pharaoh with signs and wonders (48.9–12), but there is no explicit description of the signs Moses was given to show Pharaoh (such as the rod-snake transformation or the temporarily leprous hand), just as the burning bush is mentioned only in an oblique reference at 48.2. Neither is there any mention of the miraculous provision of manna, quails and water in the wilderness (Exod. 16–17), even though in the biblical account all these took place before Moses ascended Sinai, the point at which the angelic narrator is supposedly giving his account. Admittedly there might be little point in the angel giving Moses a detailed account of events he had just participated in; nevertheless, Jubilees’ account of the exodus shows a striking tendency to abbreviate the miraculous and no inclination to embellish it.

So far as one can tell from this meagre material, Jubilees views God as the BNP here, with Moses possibly acting as MNP (the arch-demon Mastema also plays a role, but we shall return to that later). Concerning the ten plagues Jub. 48.5 starts by stating, ‘And the Lord executed great vengeance upon them on account of Israel’. It goes on to list the various plagues with which he smote Egypt, concluding ‘and upon all of their gods the Lord took vengeance and burned them with fire’. Thus far it appears that God is the sole actor, but in the next two verses (addressed, it should be recalled, to Moses) the text continues,

And everything was sent in your hand to announce [or ‘do’—see below] before it was done. And you related it to the king of Egypt before all his servants and before his people. And everything happened according to your word, ten great and cruel judgments came on the land of Egypt so that you might execute vengeance upon it for Israel. And the Lord did everything on account of Israel and according to his covenant which he made with Abraham that he would take vengeance upon them just as they had made them serve by force (Jub. 48.6–8).

Does this text then envisage God or Moses inflicting the ten plagues on Egypt, and if both, how is the division of labour envisaged? A preliminary problem is whether one can accept the conjectural emendation of ‘do’ to ‘announce’ in the first sentence of the passage just quoted, given that the Ethiopic manuscripts actually read ‘do’. If one can—and the emendation would certainly make good sense in context—then Moses’ main role here is to act as prophet, announcing to Pharaoh in advance what is about to happen. The phrase ‘you might execute vengeance’ then refers to Moses’ announcement of judgment rather than to his playing any role in bringing the plagues about, and this would then be consistent with the framing statements ‘And the lord executed great vengeance …’ (48.5) and ‘And the lord did everything on account of Israel …’ (48.8). In that case, Moses functions not as a BNP or MNP here, but simply as God’s spokesman. On the other hand, if the conjectural emendation be rejected, then one would have the apparent oddity that ‘everything was sent in [Moses’] hand to do before it was done’. Although on the face of it this is nonsense, O.S. Wintermute suggests, ‘Perhaps one could retain the Eth. Text by assuming a command to work sympathetic magic; “Everything was sent in your hand so that you might act (it out) before it was done”.’ Alternatively, James VanderKam suggests that ‘before it was done’ may reflect an underlying Greek πρὸ τοῦ ποιθῆναι, which may have been related to the first part of the sentence, thus giving the meaning: ‘Everything was sent through you, before it was done, so that you should do it.’

Wintermute’s suggestion would make Moses a BNP of sorts, though Jubilees shows little interest in sympathetic magic elsewhere. His (emended) translation (‘say’ rather than ‘do’) makes better sense in context, but requires (an albeit plausible) emendation to the Ethiopic text. One is thus left with VanderKam’s suggested translation as the one most probably correct, and this does strongly suggest that Moses is firmly envisaged as MNP for the plagues. But it would be unwise to try to deduce too much about the author’s views on miracles from this comparatively short passage, in which references to both God’s and Moses’ actions may simply have arisen from abbreviating the underlying biblical account. In that case Jubilees should be understood as attributing the plagues primarily to God, while acknowledging that Moses also had some part to play.

Yet Jubilees’ account of the Red Sea crossing attributes the miracle entirely to God (and the angel who is speaking) and makes no mention of Moses’ role whatsoever:

And I stood between the Egyptians and Israel, and we delivered Israel from his hand and from the hand of his people. And the Lord brought them out through the midst of the sea as through dry land. And of all the people whom he brought out to pursue after Israel the Lord our God threw into the middle of the sea into the depths of the abyss beneath the children of Israel. Just as the men of Egypt cast their sons into the river he avenged one million. And one thousand strong and ardent men perished on account of one infant whom they threw into the midst of the river from the sons of your people (Jub. 48.13–14).

There are two things to note about this passage. The first is the emphasis on retributive justice: just as the Egyptians ordered that the Israelite male infants should perish by drowning, so too are they drowned in large numbers. Indeed, the author seems more interested in this aspect than in the miracle as miracle, which is narrated as briefly as possible, given its importance in the Jewish tradition. The second point to note is that, despite this brevity, in common with the other texts we have examined Jubilees states that the crossing of the Sea involved the Israelites passing through as through dry land, and that the deliverance included the drowning of their pursuing enemies. As we have noted before, these are both features lacking in the Gospel sea miracles.

Earlier in the chapter, however, Jubilees does make more of Moses’ miracle-working role. At 48.2–3 the angel explains how he delivered Moses from the hands of the arch-demon Mastema, who would have killed Moses to prevent him executing judgment on the Egyptians. At 48.4 the angel continues, ‘And I delivered you from his hand and you did the signs and wonders which you were sent to perform in Egypt against Pharaoh, and all his house, and his servants, and his people.’ This parallels the trend we noted in Pseudo-Philo whereby such general statements attribute a larger miracle-working role to Moses than do the actual accounts of the miracles.

The biblical book of Genesis is scarcely replete with miracles, and given Jubilees’ cursory treatment of the Exodus miracles it comes as no surprise to find very few allusions to the miraculous in the bulk of the book of Jubilees, which follows the Genesis account reasonably closely. Jubilees 10.23–25 recounts that God came down and confused the tongues of the builders of the Tower of Babel, as in the biblical account (Gen. 11:5–8). Jubilees 10.26 adds the unbiblical detail that God sent a great wind to demolish the tower. This detail, shared with Sib. Or. 3.101–103 (cf. Ant. 1.118), is probably added not out of any particular interest in miracle, but to explain what became of the tower, thus filling in a gap in the biblical narrative. Jubilees 13.13 states, ‘when Pharaoh took Sarai, the wife of Abram, that the Lord plagued Pharaoh and his house with great plagues’ on her account. But this only repeats what the biblical account says at Gen. 12:17, and does not develop the story as does, say, the Genesis Apocryphon (on which see Chapter 7 Section 2 below), to which Jubilees is in other ways closely related. Jubilees 16.5–6 briefly narrates the destruction of Sodom and Gomorrah and threatens that other places will be similarly destroyed if ‘they act according to the pollution of Sodom’. Here the focus is once again on retributive justice. At 26.18 Jubilees perhaps introduces a minor non-biblical miracle when heaven confuses Isaac’s mind so that he mistakes Jacob for Esau and gives the former his blessing; but the purpose of this is partly to play down the deliberate deception perpetrated by two characters (Jacob and Rebecca) Jubilees otherwise portrays in the best possible light, and partly to explain how the implausible impersonation came off. Jubilees 28.12, 24 note that God closes and opens Rachel’s womb, but, as in the biblical parallels (Gen. 29:31; 30:22) the point is that God is initially recompensing Leah for Jacob’s favouritism towards Rachel, and nothing is made of a miraculous conception of Joseph per se. Likewise, Jubilees plays down any miraculous element in the birth of Isaac, perhaps in order to remove any impression that either Abraham or Sarah ever doubted God’s promises.

About the only result one might glean from this meagre material is that where Jubilees does report something that might be counted as miraculous, the agent is always God. Thus far, Jubilees appears to shows no interest in any human miracle-workers, even as minimal BNPs or PNPs, with the sole (and even then minimal) exception of Moses. But Jubilees‘ demonology is yet to be examined, and, as will be seen, this raises the possibility that Abraham and Noah are portrayed as exorcistic PNPs.

In some respects Jubilees’ redaction of Genesis resembles Matthew’s of Mark’s: many narratives are drastically abbreviated, some new narratives are added, but the main concern is in adding halakhic or paraenetic material in the form of speeches and testimonies. In this process it is perhaps only to be expected that the miraculous element might be diminished rather than enhanced (though a comparison of Matthew with Mark indicates that this need not necessarily be the case). Nevertheless, it can hardly be said that Jubilees exhibits any great interest in the miraculous as such. Indeed, it shows far less interest in miracles than Josephus, Philo, or Pseudo-Philo, and for that matter, less than either Ben Sira or the Wisdom of Solomon. On the other hand, as will now become apparent, Jubilees shows far more interest in demons than these other authors.

In Jubilees the demons are said, for example, to mislead Noah’s children and grandchildren (Jub. 7.27, 10.1), as well as blinding and killing his grandchildren (10.2). The ten per cent that remain after the rest are bound are engaged in causing illness and attempting seduction (10.11–12). Their leader, Mastema (sometimes referred to as ‘Satan’ or ‘Beliar’, if the same troublesome spirit is meant throughout), is subsequently said to send crows and other birds in the time of Abraham to rob people of the labours of sowing seed (11.11). Cruel spirits led by Mastema aid men in idolatry and warmongering (11.4–6). Indeed, idolatry is frequently connected with demons throughout the book, since demons incite people (Gentiles and wayward Israelites) to worship idols, which in turn is a form of demon-worship (1.11, 20; 15.31; 19.28; 22.17). After fleeing from Ur where he had made a bonfire of the local idols, Abraham spends a night stargazing in the vain hope of making a long-range weather forecast. On recognizing the futility of this exercise he prays to be delivered from ‘the hands of evil spirits which rule over the thoughts of the heart of man’ (12.20); but he nevertheless subsequently becomes the victim of Mastema’s attempt to test him through the sacrifice of Isaac (17.16; cf. Job 1:8–11). Mastema becomes particularly active during the exodus. At 48.2–3 the angel tells Moses that Mastema wanted to kill him to prevent him executing God’s judgment on Egypt, and at 48.9 that Mastema was aiding the Egyptian magicians against him. Jubilees 48.16–17 has Mastema rather than God harden Pharaoh’s heart and encourage the Egyptians to pursue the fleeing Israelites, but here Mastema is acting by divine permission, apparently to give God the opportunity to punish the Egyptians by drowning them in the sea (48.17). Finally, Mastema’s role becomes decidedly ambivalent at 49.2 when all his powers are sent (presumably by God—see 49.4) to slay the first-born of the Egyptians, while obediently avoiding houses whose lintels are smeared with the blood of the Passover lambs. Although here Mastema appears for once to be acting as God’s minister rather than his opponent, it may be that the author of Jubilees thought such a destructive plague to be more properly the work of an essentially evil being; perhaps this author naturally associated disease with demonic activity. Alternatively, perhaps one should not read ‘Mastema’ as a proper name here, so that what God sends to slay the first-born of Egypt are in fact his own ‘powers of hostility’.

Note the range of demonic activity in this text. Demons are responsible for disease, idolatry, leading people astray and inciting them to oppose God’s will, and also perhaps for other social and natural ills besides. Nowhere is it said that they actually possess people, either in the narrow sense of displacing their personalities or in the broader sense of invading their bodies, although it is possible that they were envisaged as causing illness through such invasions. Conversely, there is nothing that obviously resembles an exorcism in the sense of reversing such possession.

Nevertheless, Armin Lange finds material he regards as exorcistic at Jub. 12.19–20 and 10.1–14. The first of these passages follows the account of Abraham deciding it is futile to attempt weather predictions from observation of the stars (12.16). At 12.19 he sings the praises of God as creator, and in the following verse he pleas for help against evil spirits:

Save me from the hands of evil spirits

which rule over the thought of the heart of man,

and do not let them lead me astray from following you, O my God

(Jub. 12.20).

According to Lange, ‘it seems appropriate to describe Abraham’s prayer in Jub. 12.19–20 as a hymnic exorcism which banishes the evil spirits linked with astrology’. But there are a number of problems with this interpretation. First, although evil spirits ruling over human thoughts could be construed as a reference to possession in the narrow sense, in the context of Jubilees it is far more likely to refer to the demonic tendency to lead human beings astray, which is not really possession in any sense; temptation and seduction are not the same thing as possession (not that Lange claims they are). Secondly, nothing in the text suggests that Abraham is already the victim of any kind of demonic attack when he prays this prayer. The word ‘exorcism’ should at least be reserved for the cure of demonic assaults, and not extended to prevention. To use the word ‘exorcism’ to describe every type of apotropaic procedure hardly aids clarity.

At least Jub. 10.1–14 does refer to a cure, since here the demons have already begun blinding and killing Noah’s grandchildren and leading them astray (10.1–2). Noah accordingly prays to God for deliverance from these spirits (10.3–6), as a result of which God instructs his angels to bind them (10.7). Due to Mastema’s intercession, however, one-tenth remain unbound (10.8–9). Since these still represent a real threat, God orders the angels to instruct Noah in counter-measures such as healing ‘by means of the herbs of the earth’ (10.10–14).

Lange claims to find two different types of exorcism here, but then goes on to describe the second (the instruction of Noah in the use of herbs) as ‘an aetiology of medicine’. This description is fair enough, but surely undermines the claim that this second type of procedure is a species of exorcism. It is conceivable that the herbs were to be employed as material means for the expulsion of demons (like the fish guts in Tobit), but the text does not say this, and we are not entitled to assume it (neither does Lange do so). What is there, rather, is one instance of many in Jubilees of the revelation of useful knowledge to human beings through angelic intermediaries (compare, for example, the function ascribed to the Watchers at 4.15). Lange’s suggestion that in this instance the revelation serves to legitimate the use of Hellenistic medicine within Judaism is plausible (though this would be ironic given Jubilees’ staunch stand against assimilation to Hellenism in other aspects of life), but it also removes any suggestion of anything essentially magical or miraculous.

The first type of exorcism Lange identifies in this passage is another matter. According to Lange, ‘In Jub. 10.1–14 Noah banishes nine-tenths of the world’s spirits in a similar way [sc. to that of Abraham’s prayer].’ This is a misleading way of stating it, since it is not Noah, but rather the angels under God’s command who actually effect this banishing. God (assisted by angels) clearly remains the BNP here. But Lange is quite correct in observing that in Jubilees’ account, God acts in response to Noah’s prayer, so we must concede that Noah acts as PNP here. It should be noted in passing that, according to the terminology adopted in this study, this puts Noah’s activity firmly in the realm of the miraculous rather than the magical. But the more pressing question is, should Noah’s prayer be described as an exorcism?

This question is complicated by the fact that the binding of nine-tenths of the demons forms part of an account of demonic origins that Jubilees otherwise shares with 1 Enoch (on which see above). In both works angels called Watchers, struck by the beauty of human females, mate with them (1 En. 6; Jub. 5.1; 7.21). The women subsequently give birth to giants, following which there is an escalation of violence and wickedness upon the earth (1 En. 7; 9.9; Jub. 5.1–2; 7.22–24). It is this evil and oppression that causes God to cleanse the earth through the Flood (1 En. 10.1–3; Jub. 5.3–5; 7.25), and also both to command the binding of the rebel angels in the abyss pending their ultimate destruction on the Day of Judgment (1 En. 10.4–6; 12–14; Jub. 5.6, 10–11) and to cause the mutual destruction of their bastard giant offspring (1 En. 10.9–10; Jub. 5.7–9). Despite the physical destruction of the giants by mutual slaughter and flood, however, they live on in the form of the evil spirits that come forth from their bodies (1 En. 15.8–12; Jub. 10.5). This myth accounts not simply for the origin of evil spirits, but for the origin of most of the evil in the world.

Jubilees is probably dependent on parts of 1 Enoch (and perhaps other sources as well) for this account (see, for example, the account of Enoch given at Jub. 4.15–26). It differs from 1 Enoch‘s account, however, in the matter of Noah’s plea for help against the demons. 1 Enoch 10 indicates that God sends one angel (Asuryal) to warn Noah to prepare for the flood at the same time as sending another (Raphael) to bind one of the leaders of the rebel angels (Asael) and a third (Michael) to bind the rest. The imprisoned Watchers subsequently ask Enoch to intercede for them (1 En. 13), as a result of which Enoch is commanded to tell them that though they will remain imprisoned, their offspring will live on as demons to continue corrupting the earth (1 En. 15). A second binding, that is a binding of the demonic offspring, is not narrated. Noah’s postdiluvian plea and the subsequent binding of these demons (as an addition to the binding of their angelic parents) are thus peculiar to Jubilees. The binding of the demons partially assimilates the fate of the evil spirits to that of their fallen angel fathers, potentially a further blurring of the two conceptions of the demonic we noted above in connexion with the Book of Watchers (see above). If this should be combined with the suggestion that demonic evil in general can be defeated by exorcism, we might be one step closer to finding a possible context for an eschatological understanding of driving out demons.

So to return to the main question, in what sense do Noah’s prayer and God’s response at Jub. 10.1–9 constitute an exorcism? Of itself, the binding and imprisonment of demons does not necessarily constitute an exorcism, since the effect may be to prevent future demonic attacks rather than to stop present ones. To be sure, in this case some type of demonic attack is already in progress (10.1–2), but since one-tenth of the demons remain at liberty it is less clear that Noah’s prayer puts a total stop to them. Here the evidence of the text is ambiguous, since on the one hand it states, ‘And the evil spirits were restrained from following the sons of Noah’ (10.13), while on the other the previous verse describes how Noah was instructed in counter-measures that would hardly be needed if this restraint were totally effective.

If ‘exorcism’ means delivering one or more persons from demonic possession, or, by extension, from other forms of direct demonic attack, then it is perhaps misleading to call Noah’s prayer an ‘exorcism’. On the one hand it does not achieve enough, since the text does not actually state that any particular individual was delivered from demonic oppression as a result of Noah’s prayer. On the other, it achieves too much, since the normal effect of exorcism is simply to deliver one or more individuals from demonic oppression (regardless of the subsequent fate of the troublesome evil spirit), not to cause nine-tenths of the demonic population to be bound (presumably pending final judgment). To this extent, Lange is quite right to call attention to the eschatological character of Noah’s prayer at 10.5. But if Noah’s prayer is not an exorcism in the normal sense, it is clearly related to exorcism, not only in being effective in countering actual demonic attacks, but also, as will become apparent once more texts have been examined, by virtue of its form.

This result is of considerable significance as possible background to Jesus’ exorcistic activity, in which the exorcism of demons from individual sufferers is given eschatological significance (e.g. Mt. 12:28//Lk. 11:20). To be sure, what Noah does in Jubilees 10 is certainly not an exorcism in the narrow sense of reversing demonic possession as Jesus’ exorcisms seem to have been, but it does appear to provide a matrix in which a link between exorcism and the final overthrow of demonic evil could be made.

Noah’s prayer is not the only reference to eschatology in Jubilees. In Jubilees 23, following a description of Abraham’s death and burial (23.1–7) there comes a discussion of the decline of the human life-span (23.8–13), which shades into an apocalyptic section detailing the punishment and corruption of an evil generation, followed by their repentance and restoration to health and longevity (23.14–31). This thus looks forward to an idyllic future, when people will live to be a thousand years old:

And all of their days they will be complete

and live in peace and rejoicing

and there will be no Satan and no evil (one) who will destroy,

because all of their days will be days of blessing and healing.

And then the Lord will heal his servants,

and they will rise up and see great peace.

And they will drive out their enemies,

and the righteous ones will see and give praise,

and rejoice for ever and ever with joy;

and they will see all of their judgments and all of their curses among their enemies (Jub. 23.29–30)

At first sight this appears to express an eschatological hope for a time when Satan is finally defeated and God overcomes disease, a scenario that might have some points of contact with Jesus’ healing ministry in the Gospels. On closer examination, however, it is not so clear that the passage should be read this way. For one thing, in common with many other apocalypses, this one almost certainly refers largely to specific concerns at the time of the writer. In this case, the reference to the sin, corruption and covenant abandonment (23.16–17) of a future generation (future, that is, from the perspective of Moses, who is the ostensible naratee here) should probably be read as a reference to the Hellenizing party, just as the bloodletting by ruthless Gentiles (23.22–24) probably refers to Antiochus Epiphanes’ attempt to suppress traditional Jewish practices by force, together, perhaps, with the Maccabaean conflict that ensued (23.18–21). The future the writer looks forward to is one in which the Gentile oppressors have been driven out (23.30) and assimilationist Jews have returned to traditional torah observance (23.26). In this context the ‘healing’ envisaged is probably not so much the curing of diseases as the restoration of the covenant people to both independence from Gentile interference and faithfulness to the covenant.65

For another thing, the author here seems to be indulging in a certain amount of rhetorical exaggeration, so not everything he says should be taken as a literal blueprint for end-time expectations. According to Jub. 23.25, when things are at their worst, ‘the heads of children will be white with grey hairs, and an infant three weeks old will look aged’. The author of Jubilees may well believe he is living in (or has just lived through) the terrible times he describes, but presumably he has not in fact encountered many such prematurely aged babies. The expectation of vastly increased longevity should thus be treated with corresponding caution.67

Finally, the expectation that ‘there will be no Satan’ (Jub. 23.29) needs to be considered in connexion with other passages in Jubilees that say something similar. The description of Joseph’s rule over Egypt contains the notice that ‘there was no Satan and there was no evil’ on account of the virtues of Joseph’s rule (40.9). Jubilees 46.2 likewise recalls that ‘there was no Satan or anything evil all the days of the life of Joseph which he lived after his father, Jacob, because all of the Egyptians were honoring the children of Israel all the days of the life of Joseph’. Finally, in the course of a chapter spelling out the Sabbath laws, Jub. 50.5 indicates that once ‘Israel is purified from all the sin of fornication, and defilement, and uncleanness, and sin and error … they will dwell in confidence in all the land. And then it will not have any Satan or any evil (one).’ It is noteworthy that it is in precisely these contexts that the name ‘Satan’ tends to be used in Jubilees. This leads one to wonder whether in the original Hebrew text of Jubilees ‘Satan’ was actually meant to function as a proper name for the prince of demons (normally called ‘Mastema’ or occasionally ‘Beliar’ in Jubilees), or whether it simply means ‘enemy’. In each case considered so far it would make perfectly good sense to suppose that the text could have a human adversary in mind; that is, once the Gentile enemies have been driven out of the land, Israel will no longer suffer from their evil and oppression (23.29). Under Joseph, the Israelites had no enemy (and consequently suffered no evil) in Egypt (40.9; 46.2), although after Joseph’s death this would change. If and when Israel dwells in the land in full obedience to God’s command, they will be free both from evil and from foreign domination (50.5).

One might think that this interpretation was precluded by the statement at Jub. 10.11 that ‘All of the evil ones, who were cruel, we bound in the place of judgment, but a tenth of them we let remain so that they might be subject to Satan upon the earth.’ Since the one-tenth were allowed to remain free at Mastema’s request so that Mastema might have some authority left, this surely suggests that Satan and Mastema are one and the same here. But then one might well ask why the author suddenly introduces a different name for the same demonic being. One answer might be that ‘Satan’ is not a name at all, but once again the common noun ‘adversary’, which would aptly describe Mastema’s function here, as well as reflecting his name (משטמה= ‘enmity’, ‘hostility’), without prejudice to the reference of ‘adversary’ in other contexts.

If it is nevertheless felt that the association of Satan and Mastema in this chapter necessitates a demonic understanding of Satan elsewhere in Jubilees, then on the face of it Jubilees has a curious understanding of the demonic. For the text would then appear to imply that under certain conditions of particularly good human (Israelite) behaviour, Satan ceases to exist, but that he can come back into existence again when this behaviour falls below the requisite standard (that, at least, is surely how one would have to understand the notice that there was no Satan in the days of Joseph, given Mastema’s subsequent role in the exodus events). This might make perfectly good sense if talk of the demonic were seen as a mythological way of talking about human evil. Yet one may feel that to attribute conscious awareness of using mythological language about the demonic in this way to the author of Jubilees would be to make him too modern. It would also undermine the book’s value as a serious witness to Jewish demonological beliefs. On the other hand, it could be that the author of Jubilees saw the language of the demonic as a way of talking both about empirical realities and their spiritual or heavenly counterparts. Language that talks about Satan not existing might then be understood as meaning that he ceases to be effective in human affairs (at least for a while).

Perhaps the best way of tying up the loose ends left by these conflicting interpretations of Satan and Mastema is in terms of the further blurring of the distinction between the (cosmic) fallen angels and the (earthly) evil spirits that was discussed in connexion with the Book of Watchers. Not only (as we have already seen) do the majority of the demons in Jubilees 10 share the Watchers’ interim punishment (binding), but throughout the book of Jubilees they begin to take on some of the functions earlier assigned to the Watchers (in particular, leading humans astray). It may be that for Jubilees, ‘Satan’, or ‘the adversary’, names the cosmic principle of evil (derived from the fallen angels), which tends to manifest itself in the form of human adversaries who afflict Israel (in line with the social/political/moral orientation of this understanding of the demonic), whereas ‘Mastema’ is the leader of the evil spirits, whose original role was to afflict human beings individually and directly through disease and other such misfortunes. The close association, or even confusion, of Mastema and Satan in Jubilees 10 then parallels the tendency for the demons in Jubilees to take on some of the functions of the Watchers. This is a significant step towards unifying the disparate forces of evil into a diabolical army opposed to God (the conception that appears to be operative in the Synoptic Gospels). Once this step is taken, it becomes clearer how an attack on any part of the demonic army, even through individual exorcisms, could come to be seen as an attack on the forces of cosmic evil in general.

Conclusions and Developments

This chapter has examined only two examples of Enochic literature, the Book of Watchers and Jubilees. Although Jubilees contains a few accounts of miracles, neither text shows much interest in the miraculous. On the other hand, both texts contain an account of the origin and function of evil spirits that may well be important for understanding the context of Jesus’ exorcisms. The Book of Watchers appears to connect two conceptions of the demonic, the cosmic/moral and the individual/affliction, through the medium of giants, who are both the offspring of the union between the angels and human women and the source of evil spirits. In Jubilees the role of the evil spirits is increased, and begins to become assimilated to that of the fallen angels.

Paolo Sacchi’s account of the development of Jewish ideas of the devil indicates a continuation of this trend. For example, according to Sacchi, 1 Enoch 54 suggests that the Book of Similitudes has completely identified the evil spirits with the fallen angels, since for this author the binding of Azazel’s army still lies in the future. Perhaps this is to read a little too much into the text, but it is certainly the case that the description of the priestly Messiah at T. Levi 18 includes the statement that Beliar will be bound by him (T. Levi 18.12). This indicates both that the binding of Beliar is still something that lies in the future (with consequences analogous to those Sacchi finds from 1 Enoch 54), and that the binding of Beliar is now seen as a possible messianic function (if it is not stretching matters too far to call the eschatological priest of T. Levi 18 ‘messianic’ even though the text itself does not actually do so). On the other hand, the blending of the two conceptions of the demonic in the Testaments of the Twelve Patriarchs results in evil spirits that are neither cosmic-moral nor individual-afflicting but rather individual-moral. The main role of these evil spirits is now to lure individuals into sins such as lust, envy and avarice (e.g. T. Reub. 3.1–9; T. Sim. 3.1; T. Jud. 13.3; T. Dan 4.5; T. Gad 3.1), although they can be resisted by the strong-minded and those who turn to God (e.g. T. Sim. 3.5; T. Jud. 20; T. Iss. 4.4; 7.7; T. Naph. 3.1; 8.4; T. Benj. 3.2–5; 6.1). Satan/Beliar’s main role thus becomes that of chief tempter and deceiver (e.g. T. levi 3.3; 19.1; T. Zeb. 9.8; T. Dan 3.6; 6.1; T. Ash. 1.8–9; T. Benj. 7.1–2). This bears considerable resemblance to the role of the two spirits in Qumran dualism, but lacks the Qumran sect’s determinism.

It should not be supposed, however, that these trends continue in a uniform manner. On Boccaccini’s hypothesis, the later documents of ‘mainstream’ Enochic Judaism moved away from blaming evil on fallen angels towards blaming it on the fall of Adam. By the time of one of the last literary productions of Enochic Judaism, 4 Ezra, the fallen angels and evil spirits have entirely fallen out of sight, leaving innate human sinfulness as the central problem. It is also noteworthy that these texts have very little to say about miracle in general or exorcism in particular.

On the other hand, there is another branch of broadly ‘Enochic’ literature, that represented by the finds at Qumran, in which interest in the demonic persists and in which there does seem to be some interest in miracles, particularly of healing and exorcism. It is to these texts that the following chapter will therefore be devoted.

Chapter 7

Selected Qumran Texts


The two examples of Enochic’ literature discussed in the previous chapter indicated a distinctive notion of the origin of evil that could provide an important part of the context for understanding Jesus’ exorcisms. Boccaccini suggests that the Qumran sectarians were an offshoot from the group that produced this Enochic literature, and even if his theory should turn out to be not correct in every detail, the Qumran literature certainly appears to have been influenced by Enochic ideas. In looking at some of the texts discovered at Qumran the present chapter will accordingly be pursuing an ‘Enochic’ strand. Moreover, not all the Qumran texts are necessarily sectarian; indeed at least the first two of those to be examined are almost certainly pre-sectarian, and are hence best understood as further examples of literature from the general milieu of 1 Enoch and Jubilees.

Once again it will be necessary to change the manner of treatment, since virtually all the texts to be examined in this chapter are both brief and fragmentary. On the one hand this means that we shall be looking at quite a number of texts instead of focusing on just one or two, while on the other it will become necessary to discuss some detailed problems of reconstruction and interpretation. Whereas the overall aim remains the same, to investigate the Second Temple context of Jesus’ miracles, this may result in a perceived shift in approach.

Although the majority of scholars take the Qumran sectarians to have been Essenes, there are some dissenters. Even if their arguments do not exclude the possibility that the Qumran sectarians were Essenes they raise serious methodological questions about using the classical sources (Philo, Pliny and Josephus) to interpret the Qumran texts. Attempting to harmonize the classical sources with the Qumran texts may well result in a false conflation of diverse elements. Methodologically it is therefore preferable to treat the classical sources about the Essenes and the Qumran documents as two separate sets of texts, to be interpreted apart from each other and only compared once one has established what picture is presented by each group separately. This chapter will be concerned with the Qumran texts, not with the classical sources on the Essenes.

The Qumran literature does not display much interest in miracle, but there are occasional references, and a number of texts that merit closer examination. Three texts that have received particular attention in relation to the Gospel miracles are the Genesis Apocryphon (1QapGen), the Prayer of Nabonidus (4QprNab = 4Q242) and the Messianic Apocalypse (4Q521). There are also a number of texts potentially related to demonology and exorcism (e.g. 4Q510, 4Q511, HQPsa, 11Q11, 4Q560 and 4Q266). But before discussing all these texts in detail there are a couple of general points that should be made in passing.

The first is that the Qumran literature contains virtually nothing that looks like a ‘nature miracle’. There is a brief reference to the Red Sea miracle at 1QM 11.10, in the context of recalling God’s past interventions as a ground of hope that he will do something similar in the great eschatological war to come. There is also a general admonition to ‘Remember His miracles which he did in Egypt’ at 4Q185 1.10. But it is in relation to healing and exorcisms that the most promising connexions between the Gospel miracles and the Qumran literature are to be found, and it is this area that has attracted most scholarly attention. There are one or two places where the Qumran texts mention signs and wonders in general (e.g. 1QS 11.23; 4Q504 2.12; 6.1–11; 4Q392). In none of these texts, however, is it clear that the words translated ‘marvels’, ‘signs’ and ‘wonders’ denote miracles, as opposed to God’s actions in the history of Israel, first calling them to be a people and then disciplining them through the vicissitudes of their collective experiences.

The second is that in spite of the attention given to the Essenes as healers (e.g. War 2. 136, see Chapter 2 above), there is little in the Qumran texts about healing physical ailments. Some texts (e.g. Rule of the Community 4.6; Damascus Document 8.1–5; Hodayot 2.8–9; 9.24–25) do refer to healing as an eschatological hope, but the notion may be one of ‘spiritual healing, the cure of sin’ rather than the healing of illness. Likewise the hope expressed for the healing of blindness in 4Q504, column 2 is most probably to be understand as a healing of spiritual blindness. The wonders hoped for a few lines earlier in this text are more likely to be miracles of salvation like those of the exodus than healing miracles.8 Nonetheless, there are some Qumran texts that do appear to concern healing and exorcism more closely, and it is these that will now be examined.

The Genesis Apocryphon (1QapGen)

The Genesis Apocryphon is preserved in a scroll from Cave 1, of which 22 columns survive, all but five in fragmentary form. The surviving text begins with an account of Lamech’s suspicions concerning the conception of his son Noah. In particular he fears that his wife, Bathenosh, has become pregnant by one of the Watchers (and therefore, presumably, that she is about to give birth to a giant). Despite Bathenosh’s protestations of innocence, Lamech asks his father Methuselah to ask his father Enoch what the truth of the matter is. The text breaks off at the point where Methuselah begins to put his question to Enoch. The connexion with the Enochic traditions explored in the previous chapter is thus clear. Although this section of the text adds nothing to our knowledge of Enochic demonology, it does thus serve to locate the Genesis Apocryphon within the same general frame of thought. Moreover, a demon does feature later on in the text.

The text is fragmentary for the next several columns until the story of Abraham, which survives in some detail. The passage of interest occurs at the point when Abraham journeys down into Egypt during a famine and tries to pass Sarah off as his sister.

The account of this incident in the Genesis Apocryphon (1QapGen 19–20) follows the same general outline as the biblical account in Gen. 12:10–20 (with important variations to make Abraham appear more innocent and Pharaoh more guilty) but adds a considerable detail (not least in the description of Sarah’s beauty). Of particular interest is the way these additions turn Abraham from a largely passive observer of events into a PNP. According to this version, Abraham is warned in a dream that he might get into trouble on account of Sarah. He spends his first five years in Egypt undisturbed, until three princes are sent from the king to enquire after his business. They return to Pharaoh and praise Sarah’s beauty, whereupon Pharaoh has Sarah forcibly abducted so that she might be his wife, and only spares Abraham because Sarah tells him he is her brother. Abraham and Lot spend the night weeping, and Abraham implores God not to allow Pharaoh to defile Sarah but to raise his mighty hand against Pharaoh and his household. God duly sends a pestilential spirit to afflict Pharaoh and his household, with the result that the king is unable to have intercourse with Sarah, despite being with her for two years. After two years his illness becomes even more severe, so Pharaoh sends for all the magicians and physicians of Egypt. But they prove unable to effect a cure, since the evil spirit afflicts them too. One Hirqanos then come to Abraham and asks him to lay his hands on the king and cure him, but Lot explains that his uncle cannot do this while the king is still in possession of Abraham’s wife. Hirqanos duly reports to the king and bids him return Sarah to her rightful husband if he wants Abraham to pray for him. Pharaoh thus summons Abraham, complains of his falsely passing Sarah off as his sister, and banishes them both from Egypt. He then asks Abraham to pray for him, that the evil spirit may be rebuked. Abraham accordingly prays and lays his hands on the king, and the king recovers, the evil spirit being duly rebuked.

It is significant that Abraham is represented as laying his hands on Pharaoh. Some time ago, David Flusser called attention to this:

The laying-on of hands for healing purposes is not found in the Old Testament, nor in rabbinical literature (as far as we know), but it appears many times in the New Testament in the stories of healing related there of Jesus and his disciples.

From this Flusser deduced both that healing by the laying-on of hands was already being practised in Jewish circles from before the time of Jesus and that the Genesis Apocryphon supplies us with the very Aramaic word Jesus and his disciples used to describe it: סמך. In the lxx this word is translated by ἐπιτιθέναι, which is the same Greek word used for the laying-on of hands in the New Testament.

Flusser’s argument, which is not explicitly stated, is presumably that since healing by the laying-on of hands does not occur elsewhere in Jewish literature, and since it does not come from the biblical account, and since the idea must have come from somewhere, it can only have come from the author’s experience of current healing practices. Given the tendency in re-written bible stories such as the Genesis Apocryphon to bring the narrative up to date in order to ensure its continuing relevance, such a line of argument is certainly plausible.

There is a possible reference to healing by the laying-on of hands in the lxx of 2 Kgs 5:11. The Hebrew phrase והנירּ ידו (‘and wave his hand’) is rendered by the lxx as ἐπιθήσει τήν χεῖρα αὑτοῦ, ‘and will lay on his hand’. This apparently undermines Flusser’s case that no other Jewish text of the period refers to the laying-on of hands for the purposes of healing. But the fact that the translators of the lxx chose to render ‘wave [or ‘lift up’] his hand’ by ‘lay on his hand’ could well be taken as an indication that they, too, were updating the text to reflect current healing practices.

Whereas in the Gospels exorcisms and healings tend to be seen as separate activities, in the present passage the categories are blurred; the healing is effected by exorcism. Note, however, that the evil spirit in the Genesis Apocryphon is not driven out of a possessed victim, but rather driven away from a sick victim; at no point in the story is the king’s personality said to have been displaced by that of an invader. He is instead afflicted with some ill-defined sickness, which, inter alia, prevents him from consummating his ‘marriage’ to the kidnapped Sarah. As Vermes usefully comments on this passage,

The description corresponds to an image of a spirit armed with a stick, whose business it was to keep every man away from Sarah. It even put to flight the approaching magicians.

Rabbinic tradition gives an analogous interpretation of this verse:

R. Berekiah said: Because he dared approach the lady’s shoe. R. Levi said: An angel stood there all night long with a whip in his hand. When Sarah said to him ‘Strike’, he struck. When she said ‘Cease’, he ceased.

A further difference between this Abraham story and the Gospel stories is that in the New Testament exorcism is never accompanied by prayer. The only apparent exception is at Mk 9:29. But Mk 9:29 does not say that Jesus prayed, but only that he advised his disciples to do so. As Mark tells this story, Jesus would have had very little time for prayer. The twofold pattern of praying and laying-on of hands is found at Acts 28:8, but here the healing is carried out not by Jesus but by Paul. Thus neither example proves to be an exception to the rule that Jesus’ healings and exorcisms were not carried out through prayer.19 The Abraham story of 1QapGen represents Abraham as a PNP and MNP whereas the Gospels represent Jesus as a BNP.

There is yet a far more significant difference between this story and exorcisms in the Gospels. In the Gospels the evil spirits act on their own initiative in opposition to God’s rule. The evil spirit in the Genesis
Apocryphon, on the other hand, acts at God’s behest to protect Sarah. The illness that Abraham ‘cures’ as PNP and BNP is one he first causes (or at least requests) as PNP. Moreover, so far as we can tell from the Genesis Apocryphon this is the only time Abraham acts thus; there is nothing to suggest that he exercised a gift of healing on a regular basis.

That said, the spirit that afflicts Pharaoh is described as an evil spirit in a text that appears elsewhere to stand in the tradition that evil spirits originated from the bodies of the gigantic offspring of fallen angels and human women. One may thus wonder how it is that this evil spirit appears to be acting in God’s service. But we have encountered something of the sort before: in Jubilees Mastema is said to have despatched all his powers to slay the first-born of the Egyptians at the time of the Passover (see Chapter 6 Section 3 above). This was a destructive act, but it was within God’s plan for the liberation of the Israelites. Moreover, the Genesis Apocryphon may also have borrowed from the exodus story in having Pharaoh summon all his magicians, to no avail. If the author of the Genesis Apocryphon was deliberately colouring the Pharaoh of his story with the shades of the Pharaoh of the exodus, and if he knew a tradition that the first-born of Egypt had been slain through the agency of evil spirits, as Jubilees suggests, then this might account for the seemingly helpful evil spirit here.

The differences from the Gospel exorcisms (and healings) remain, but there are also similarities. The evil spirit of the Genesis Apocryphon does not possess Pharaoh, but it does afflict him, and thus illustrates the sort of thing evil spirits might be expected to do. Again, even though in the story Abraham is only asking God to reverse the affliction God sent to protect Sarah at Abraham’s request, the procedure Abraham adopts to rid Pharaoh of the evil spirit could well reflect current methods of countering demonic attacks.

The Prayer of Nabonidus (4QprNab)

The Prayer of Nabonidus is another Aramaic text. It appears to contain the prayer of one Nabunai or Nabnai (נבני) who is plausibly identified (by references within the text) with Nabonidus, the last king of the Babylonian Empire. In this document Nabonidus describes how, in the course of an illness, he confessed his sins to the true God, who duly granted him a Jewish exorcist or soothsayer.

The text survives only in very fragmentary form. A photograph of the surviving fragment appears facing page 408 of the Revue Biblique of 1956. This picture shows one reasonably large fragment containing the start of eight lines, together with three smaller fragments. The arrangement of the fragments in the photograph, together with the transcription on the facing page, indicate that J.T. Milik has taken two of the smaller fragments, numbered 2 and 3, to contain partial continuations of the lines begun in the main fragment, numbered 1. In this he has been followed by Pierre Grelot in his ‘new attempt at restoration’ of this text. Nonetheless, fragments 2 and 3 do not appear to join directly on to fragment 1, which leaves a substantial lacuna to be filled in on each line by informed guesswork. How this affects the reading of the text may be seem for example, by comparing the different transcriptions effected by Milik and Grelot.

One thing these transcriptions show is the extent of the lacunae that have to be filled to make sense of this text. This possibility of reconstructing the text in various ways leads to some variety in translation. This becomes apparent when, for example, one compares the French translations of Milik and Grelot, together with the English translation of Vermes.

Although differing from one another on several details, not all of them minor, these three translations nevertheless agree on the same broad outline. The piece is introduced as the words of the prayer spoken by Nabunai, king of Babylon, when he was suffering from an ulcer at the decree of the Most High God. The king had been suffering for seven years when he encountered a Jewish exorcist/soothsayer who pardoned his sins (or else God forgave the king’s sins, or else the king confessed his sins and God sent him the Jewish soothsayer). This Jew commanded the king to recount his experiences in writing to the glory and honour of the Most High God. What follows is then either what the king wrote on the Jew’s command, or what the Jew wrote to the king. Either way it narrates how the king prayed to the gods of gold and silver and bronze for seven years thinking them to be gods. It is assumed that the king then saw the error of his ways, repented of his sins, and was duly healed by the Jew. There thus appears to be a healing story, and one that appears to link healing and forgiveness.

This reconstruction relies on the assumption that the Prayer of Nabonidus follows roughly the same outline as the story of Nebuchadnezzar’s indisposition in Daniel 4. This may be reasonable, but it is worth observing what the text actually says without the conjectural completion of the lacunae. To this end I give Vermes’s translation with the lacunae left blank:

The words of the prayer uttered by Nabunai king of the 1[…]bylon, […] king, […] with an evil ulcer in Teiman by decree of the […].

I was afflicted […] for seven years … and an exorcist pardoned my sins. He was a Jew from […], ‘Recount this in writing to […] the name of the […]

‘I was afflicted with an […] ulcer in Teiman […] For seven years […] prayed to the gods of silver and gold, […] wood and stone and clay, because […] that they were gods.

Note that this fragment does not actually say that the Jew was from among the exiles (like Daniel), nor does it actually mention the ‘Most High God’ (imported, no doubt, from Dan. 4:2) nor does it even say that the king was healed. Moreover the translation of the phrase ‘an exorcist pardoned my sins’ is far from certain. Milik has ‘[when I had confessed my sins and] my faults, [God] granted me a soothsayer’, whereas in Grelot’s view it is God who does the forgiving (and healing). One assumes that a fragment preserved in a Qumran cave would be likely to express broadly ‘orthodox’ views from a Jewish standpoint, but this is only an assumption: so far as the surviving text is concerned, the Jew might be an apostate from Yahwism, and Nabunai a convert to the worship of Sin the moon-god, as in historical reality. Or if one fills in the final lacuna with the words ‘I knew’ the text might go on to say ‘and after seven years the gods of silver and gold hearkened unto my prayer and I was healed’. To be sure, this is unlikely, given the occurrence of the similar expression ‘and you have praised the gods of silver and gold, of bronze, iron, wood, and stone, which do not see or hear or know’ at Dan. 5:23, but then, given the possibility that the author of Daniel was familiar with this text and made use of it (see below), it could be that Daniel 4–5 contains a polemical adaptation of this text which (perhaps) did not originally aim to praise the God of Israel.

Nonetheless, the reconstruction of the Prayer of Nabonidus along the lines suggested by Daniel 4 certainly seems plausible. It is thus eminently reasonable when A. Dupont-Sommer surmises:

It is easy to suppose that, in the sequel, the king, after having invoked the false gods in vain, addresses his prayer to the true God, to God Most High: the text of this prayer of Nabonidus, which must constitute the first part of our writing, thus the indication of the title of the scroll, has entirely disappeared; likewise the detailed narration of the healing performed by the Jewish exorcist has disappeared, a narrative which must have followed the prayer of the king.

Dupont-Sommer has some textual basis for supposing that a healing narrative ‘must have’ followed the (supposed) prayer of the king, since there is one further fragment of the text (insufficiently preserved to make much sense) which is yet to be considered here. Dupont-Sommer thus continues:

A fifth small fragment, isolated, presents the remains of five lines. On the first line of this fragment one reads this: ‘… and without them I was healed’ [et sans eux je fus guéri]; let us understand, no doubt: without the aid of the false gods.

This, however, involves us in further complications, since Milik’s translation of this line reads not ‘and without them I was healed’ but ‘and in addition, I had a dream’ (Et en plus, j’ai eu un songe). Once again, the interpretation of the text is dependent upon conjecture and uncertainty, since it would in any case be extremely hazardous to guess where this fourth fragment fits into the narrative. Rather than becoming enmeshed in such speculation, I shall focus on what for my purposes are the principal cruces interpretum: the identity of the Jew in line 4 and the nature of what passed between him and the Babylonian king.

On the basis of the resemblance between the present text and Daniel 4, several writers identify the Jewish גזר with Daniel. This identification can hardly be established from the surviving fragments, and presupposes that ‘The composition is inspired by the story of Nebuchadnezzar in Daniel 4.’ The dependence, however, could easily be the other way around.34 It may be that the author of Daniel adapted this story by substituting the better known Nebuchadnezzar for the less famous Nabonidus, and the more famous Daniel for some unknown Jew. The exact nature of the relationship between the Prayer of Nabonidus and the book of Daniel is in any case unclear. Given the inconclusive nature of these arguments, the identification of the anonymous Jew of 4QprNab 4 with Daniel is far from secure. It thus seems preferable to conclude that the text concerns an anonymous Jewish גזר.

The function of this anonymous Jew is another question. It is far from clear what passes between him and the king, and equally unclear what is meant by designating him as a גזר. Milik and Grelot take the word to mean ‘devin’ (i.e. ‘seer’ or ‘soothsayer’), whereas Vermes, following Dupont-Sommer, favours ‘exorcist’. These are not the only suggestions.

The problem arises from the fact that the word גזר appears nowhere else in surviving literature, apart from here and in the Aramaic section of Daniel. In Daniel (2:27; 4:7; 5:7, 11) the גזר appears as one item in lists that also include some combination of wise men, magicians, enchanters, and Chaldeans. On each occasion the RSV translates the word גזר as ‘astrologer’, but this must be regarded as conjectural. Greek translations of Daniel, including the lxx, unhelpfully render it as γαζαρηνοί. Other ancient translators venture ‘sacrificers’, ‘mages’, ‘astrologers’ and ‘enchanters’. Dupont-Sommer arrives at the meaning ‘exorcist’ by a roundabout derivation from the base meaning of the Semitic root גזר (‘to cut’, ‘to divide’ and hence also ‘to decide’), although he notes that others have derived the meaning ‘astrologer’ from the same root. Vermes suggests that ‘because the root from which the term derives means “to decree”, a gazer is one who exorcises by decreeing the expulsion of the devil’. But the root meaning ‘decree’ or ‘decide’ would also support the translation ‘seer’ or ‘soothsayer’ in the sense of a ‘determiner of fates’ (‘détermineur de sorts’). This does not rule out the translation ‘exorcist’, but it makes it hard to prove.

Given the inconclusive nature of these arguments from etymology, one might do better to argue from context. The passages in Daniel where גזר occurs hardly suggest that an exorcist is meant. An exorcist is one who drives out or drives away evil spirits. At Dan. 2:26 Nebuchadnezzar asks Daniel to tell him his dream and supply its interpretation, to which Daniel replies that ‘No wise men, enchanters, magicians, or גזריו can show to the king the mystery which the king has asked’. The גזריו appear in similar company at Dan. 4:7; 5:7, 11. In two of these cases the גזריו are part of a group summoned to offer the interpretation of a dream, and in the other case, referred to twice, they are to lend their assistance to the interpretation of a strange writing. In none of these four instances is there any suggestion that an evil spirit needs to be cast out or chased away, neither is it at all clear why people whose expertise lay in disposing of evil spirits should be summoned to interpret either dreams or writing. In all four cases, the translation ‘seer’ or ‘astrologer’ would fit the context far better.

This also appears to be so in the situation described in the Prayer of Nabonidus. Since Nabonidus is not said to be suffering from an evil spirit, it is not clear how an exorcist would benefit him. To be sure the text contains so many lacunae that the odd demon or two could easily be lurking in one of them. Moreover disease and demon-affliction were often linked. It is therefore by no means inconceivable that a person suffering from some kind of inflammation or ulcer should be represented as gaining relief through an exorcist. But the link between disease and demon-affliction was by no means universal (in Tobit, for example, the two are kept quite distinct. Sarah is afflicted by a demon; Tobit’s blindness is ascribed to natural causes [bird-droppings]). It would thus be gratuitous to import demons into a situation where they are not explicitly mentioned unless there are good grounds for doing so.

This still does not settle what the Jewish גזר actually does for the Babylonian king. The difficulty is compounded by the mutual dependence between what one thinks the word גזר means and one’s interpretation of the context in which the troublesome word stands.

In Dupont-Sommer’s version, for example, after the king prays to the Most High God (for reasons unstated), a Jewish exorcist forgives his sins and tells him to recount his experiences to the glory of the Most High God. Lines 6 to 8 of the text then contain the start of Nabonidus’s narration. On Grelot’s reconstruction, however, it is God who makes the first move by turning his face to the king (after seven years), healing him and forgiving his sin. The function of the Jewish seer is then to instruct the king to write an account of what has happened, to the glory of God, and to issue it to his people. Lines 6–8 then contain the start of the Jewish seer’s explanation to the king of what has happened to him and why.44

It is Grelot who makes the stronger case here. If one is looking for parallels with the book of Daniel, then having Nabonidus issue a letter to his subjects would correspond with Nebuchadnezzar’s open letter in Daniel 4, and the Jewish גזר now functions in the Prayer of Nabonidus much as the eponymous hero of the book of Daniel does, as a seer explaining God’s dealings to the Babylonian ruler. Moreover, on Grelot’s interpretation the events narrated follow a logical sequence: Nabonidus is struck with an ulcer (as a punishment for idolatry, though he does not know this at the time); after seven years God mercifully forgives his sin and heals his sickness, but it is still necessary that this should be explained to Nabonidus; this is the function of the Jewish seer, who also commands the king to inform his subjects of his conversion to belief in the one true God. Apart from all the other objections to Milik’s translation, according to the sequence of events he describes, Nabonidus confesses his sins (to whom?) before God sends him a Jewish seer to explain everything, so that it is left unclear how the king knew he had any sins to confess (presumably he did not regard idolatry as sinful during the seven years he was practising it). A similar objection applies to Dupont-Sommer’s reconstruction, in which the king is made to pray to God Most High before the Jewish exorcist appears on the scene to forgive his sins. What changed his mind? On Grelot’s reconstruction the synchronicity of the healing and the Jewish seer’s explanation would suffice to convince the king of the error of his former ways, but on the other two reconstructions the king’s change of heart is left unexplained.

One aspect of Grelot’s reconstruction that is open to question is his introduction of ‘et il me guérit’ (‘and he healed me’) into the lacuna at the end of line 3 since, as he is well aware, this can only be conjectural. Grelot argues first that the healing and forgiveness are usually associated. Clearly, they are sometimes, but if too much weight is placed on this argument, the Prayer of Nabonidus cannot be used as further evidence for the link between healing and forgiveness without vicious circularity. Grelot’s second argument is that the end of line 3 needs to introduce a turning-point in the story. This is more or less the argument I have already put forward: something is needed to make the king heed the Jewish seer, and his being healed at this point in the narrative meets the case. Otherwise, if God had merely forgiven his sins without somehow demonstrating that forgiveness, how would the king know? Moreover, since the story talks about a sickness that lasted seven years, it seems reasonable to suppose that something happened to change or alleviate the sickness at the end of the seven-year period. The mention of an affliction at the opening of the story more or less requires that something shall happen to or on account of that illness as a resolution of the plot. Given these considerations, Grelot’s reconstruction seems justified here, with the caveat that it can, of course, never be more than reasonably probable.

If this reconstruction is correct, then the Jewish גזר is neither a PNP nor an MNP, since God acts directly as a BNP. To be sure, Grelot’s reconstruction is only conjectural, so there may be an account of a healing by a Jewish גזר here, but it seems more probable that there is not. In that case, the Prayer of Nabonidus does not provide any direct parallel to the healing stories of the Gospels. In particular, it provides no clear parallel to the link between healing and forgiveness supposedly exhibited in the healing of the paralytic (Mk 2:1–12 and parallels). It may, however, provide a further example of a healing story, and one in which God acts directly as BNP.

The Messianic Apocalypse (4Q521)

Several fragments of the text of the Messianic Apocalypse survive, but the one that has attracted the most attention is the largest and best preserved, that from column 2, since it appears to look forward to a Messiah who will heal the wounded, revive the dead and bring good news to the poor, and also refers to God as liberating captives, restoring sight to the blind, and straightening the bent. Although such phrases are more or less Old Testament quotations (from Isa. 61:1 and Ps. 146:7–8), their association with the coming of the Messiah makes them appear close to Gospel passages such as Lk. 4:18–21 and Mt. 11:4–5. The apparent link is strengthened by the fact that, unlike the biblical passages to which they allude, both 4Q521 and the Gospel passages refer to raising the dead.

The English translation given below is based on that of Vermes [V], adjusted in Line 8. This has been compared with several alternative translations (of Puech [P], Eisenman and Wise [E-W], and Wise and Tabor [W-T]).

1.    … [the hea] vens and the earth will listen to His Messiah,

2.    and none therein will stray from the commandments of the holy ones.

3.    Seekers of the Lord, strengthen yourselves in his service!

4.    All you hopeful in (your) heart, will you not find the Lord in this?

5.    For the Lord will consider the pious (hasidim) and call the righteous by name.

6.    Over the poor his spirit will hover and will renew the faithful with his power.

7.    And he will glorify the pious on the throne of the eternal Kingdom.

8.    liberating the captives, restoring sight to the blind, lifting up the b[ent]

9.    And f[or] ever I will clea[ve to the h]opeful and in His mercy …

10.    And the f[ruit …] will not be delayed for anyone

11.    And the Lord will accomplish glorious things which have never been as [He …]

12.    For He will heal the wounded, and revive the dead and bring good news to the poor

13.    … He will lead the uprooted and knowledge … smoke(?)

Before one can discuss the significance of this text in relation to the miracles of Jesus, one must first resolve a number of uncertainties in the translation.

Line 8: This line is a quotation from Ps. 146:7b–8. The wording is identical to the Hebrew text of the psalm verse, apart from the threefold omission of the name יהוה as the subject of each verb. If line 8 is regarded as a separate sentence, then a translation in the present tense as in Psalm 146 might be more appropriate, but this would be problematic since the participles would then have no subjects (the name YHWH having been removed in each case). The V translation (‘He who …’) seems odd, both because this would require a definite article in front of the participles (which is not there), and because it weakens the grammatical link of this line with its context. The most literal translation of the last clause of this line would be ‘lifting up the bent’. If ‘bent’ is taken literally, then the text envisages healings comparable to Lk. 13:10–13, as P and V suppose; if ‘bent’ is meant as a sociological metaphor, then the meaning is ‘downtrodden’ (as in E-W and W-T). After ‘liberating prisoners and opening the eyes of the blind’ either sense seems possible, but if the Qumran writer intended the biblical quotation as a gloss on the previous line, the sociological sense is more likely to be the one intended.

Line 10: This line is in a very poor state of repair. It has been variously translated: ‘And the frufit of a] {good} [wor]k will not be delayed for anyone’ (V, P); ‘and [His] Good[ness …] of Holiness will not delay …’ (E-W); ‘a[nd in His] go[odness forever. His] holy [Messiah] will not be slow [in coming]’ (W-T). These translations result from very different ways of reconstituting the text. The two relevantly different reconstructions of this line of Hebrew are as follows:

ופרִ[י מצש]ה טוב לאיש לוא יתאחר    (P)

וט[ובו לצד משיחו] הקדש לוא יתאחר [לבוא]    (W-T)

W-T’s introduction of the word ‘Messiah’ into this line is purely speculative. It is introduced because W-T suppose that the text is going on to talk about the Messiah in lines 11–13, but this is something yet to be established. P’s reading of the Hebrew text includes letters after the first lacuna that are incompatible with W-T’s speculative משיחו, although P marks them as damaged letters whose reading is uncertain.

Line 11: This is also a contentious line. E-W has ‘And as for the wonders that are not the work of Lord, when He …’ W-T adds ‘(i.e. the Messiah) [com]es’. V has ‘And the Lord will accomplish glorious things which have never been as [He …]’. The two important questions here are, first whether it is ‘wonders’ in the sense of miracles here, or simply ‘glorious deeds’; and second, whether the point is that the deeds will be performed by someone other than God (as in E-W and W-T) or that God is going to perform unprecedented actions (as in V and P). The differences arise partly from different ways of transcribing the Hebrew text:

ונב] דות שלוא היו יצשה אדני באשר רִ[בר]    (P)

ונב] דותִ שלוא היו מצשה אדני באשר י[בוא]    (W-T)

At Ps. 87:3 RSV renders נבבדות ‘glorious things’. It derives from the root בבד, ‘to be heavy, weighty, severe, numerous, considerable, renowned’. P’s ‘glorious deeds’ is thus to be preferred to W-T’s ‘wonders’, which reads too much into the text. Moreover, W-T achieve their translation by reading מצשה for P’s יצשה, thereby turning the imperfect verb into a noun. The phrase מצשה אדני could indeed mean ‘work of the Lord’, which would then yield the translation ‘And the glorious deeds that were not the work of the Lord when …’ Compared with E. Puech’s translation, this seems grammatically clumsy; one might expect the conjunction ‘when’ to follow a clause with a main verb (which W-T’s translation lacks), and ‘as for’ has to be understood to make sense of the W-T translation at all. On balance, Puech’s reading seems the more plausible: it is grammatically smoother, while W-T show suspicious signs of wanting the text to provide a ‘disjunction’ at this point.

Line 12: Here, V and P have ‘wounded’ (P blessés [à mort]) where E-W and W-T have ‘sick’, which seems closer to the idea of healing expressed in the Gospels. The Hebrew word in question is חללים, ‘pierced’, ‘killed’, ‘slain’, so perhaps, ‘mortally wounded’ (as in P). Although the root חלל more often means ‘kill’ than ‘wound’ in the Old Testament, the latter meaning is occasionally found (e.g. Judg. 9:40; Job 24:12; Ps. 69:26; Ezek. 26:15) and it would be odd to say ‘he will heal the slain’.

To move on to the significance of this text, the most important crux is who performs the miracles in line 12: the Messiah, or God?

If Puech’s translation of Line 11 is accepted then the context would seem to demand that it is God who performs these eschatological acts, for lines 11 and 12 then read:

And glorious actions which have never been the Lord will realize when he [speaks?] for he will heal the mortally wounded, revive the dead, and preach the good news to the poor.

Further support for seeing God as the one acting here comes from another fragment of this text, in which it appears to be the Lord who is envisaged as raising the dead. Moreover, although there are close correspondences to the wording of Isa. 61:1 in lines 8 and 11, as Eisenman and Wise point out, there are also

word-for-word correspondences to the Eighteen Benedictions, among the earliest strata of Jewish liturgy and still a part of it today: ‘You will resurrect the dead, uphold the fallen, heal the sick, release the captives, keeping faith with those asleep in the dust …’, referring obviously to God.

The correspondence is yet another reason for taking God to be the subject of the verbs in line 12 of the text.

But there is another side to the argument, since it is hard to see how the preaching of good news to the poor can be carried out directly by God, especially since at Isa. 61:1, which the text appears to echo at this point, it is expressly said to be the task of the one anointed by the spirit of the Lord for that very purpose.

The precise interpretation of line 8 becomes crucial here. If line 8 refers to God’s future actions (continuing the tense of line 7), why should God who is going to liberate captives, restore sight to the blind and straighten the bent leave healing the mortally wounded and reviving the dead to his Messiah? One would naturally expect these actions to be performed by the same agent. In which case it must be God who is the subject of the verbs in line 11, whatever difficulty may be involved in his preaching the good news to the poor. But if the author had intended the verbs in line 8 to describe the future acts of the Lord, why did he not change them into the imperfect tense? It cannot be simply because he wished to preserve the wording of Psalm 146; had that been his main concern he would either have kept יהוה as the subject of each verb, or at least replaced it with אדני instead of dropping it altogether. Besides, the change from participle to imperfect would be no greater than the change he has made to his (apparent) quotation from Isa. 61:1 at the end of line 12, changing ‘to preach good news’ to ‘he will preach good news’ in order to fit the new grammatical context. This, presumably, is why Vermes chose to translate the participles by ‘he who … ‘, understanding the author to be referring to God’s habitual acts rather than his future ones, as in the psalm from which the quotation is taken. On this reading, it could be argued that he who regularly does these activities in the present may yet plan to do his greater saving actions through his Messiah in the future.

In favour of the future interpretation is the fact that the immediate context implies a future hope, and the idea of being about to liberate captives and lift up those bowed low would constitute some kind of parallelism with the ideas expressed in the previous line. If the Lord is already liberating captives, restoring sight to the blind and elevating the downtrodden, it is unclear why the audience have to be exhorted to hold on for future deliverance; whereas if the participles are being used to express an action in the imminent future (which would be one grammatical possibility), this would tie in with the absence of delay apparent in line 10. On balance, then, it seems most likely that line 8 does intend a future reference, although the grammatical construction is admittedly unclear.

This leaves the problem of how God can be said to be about to preach the good news to the poor. Whatever weight is placed on the particular use of בשר at Isa. 61:1, nowhere in the Old Testament is this verb used with God as its subject, but always of human proclaimers of (mostly good) news. Added to the fact that Isa. 61:1, which appears to be in view here, speaks of one anointed by God’s spirit to perform various actions on God’s behalf, the difficulty seems almost insurmountable. There thus appears to be an exegetical deadlock. Every other consideration points to God being the subject of the verbs in Line 12, and yet the third of these verbs, preaching the good news, stubbornly resists having God as its subject.

At this point one must step back and consider some broader questions about the genre and purpose of this text. It appears that 4Q521 is hymnic in type. In Puech’s view, the different themes evoke the genre of an exhortation on the blessings and punishments that God will bring about in the days of his Messiah. In language that is half-prophetic and half-apocalyptic the author invites the just to persevere in the law and in the orthodox practice of the cult. This view seems reasonable. But this means this text is not a systematically constructed theological treatise but a poetic evocation of God’s imminent saving act; it is highly allusive in nature, making frequent use of psalms and prophecy and weaving them into a tapestry designed more with evocative than with didactic goals in mind. The author may not have taken minute pains to express himself precisely at every turn; he was more concerned with inspiring his audience.

This does not mean that he will have been content to write any old nonsense. But it allows the possibility that he was prepared to slip a little carelessly between one subject and another. In particular, he may not have been greatly concerned to distinguish between what God was going to bring about directly and what God was going to effect through the person of his Messiah. Or he may have considered that the action of a Messiah sent by God was the equivalent of God acting himself (on the shaliach principle). In Isa. 61:1–2 the prophetic figure (perhaps reinterpreted as the Messiah by the author of 4Q521) is anointed with God’s spirit to act as God’s agent; on the shaliach principle the acts he performs while carrying out this mission may also be seen as God’s acts: he proclaims the good news on God’s behalf, so that his words may be regarded as God’s words just as the prophets of old certified their proclamations with ‘Thus says Yahweh.’ Or again, the text may describe what God is going to do quite apart from the Messiah.

Where does this leave the raising of the dead and the other miraculous deeds? In the end, one can only say that the text does not make it clear whether these are to be performed through the Messiah or not. This is not a distinction the author was concerned to make: in common with several other authors of intertestamental texts his interest lay not with the person of the Messiah but with what God was going to do in the Messianic age. The Messiah will come and the great age of salvation will dawn (for the pious); that is the author’s message; demarcating a precise division of labour is not his concern.

Moreover, it is not even clear that the text looks forward to the performing of actual individual miracles of healing and raising the dead. As in 1.8 so also in ll. 11–13 the language may be rather the traditional language of salvation. What may be in view is not so much a literal revival of the dead or healing of the mortally wounded as the revival of God’s hard-pressed people (cf. Ezek. 37:1–14; Hos. 6:2). This may well be the language of eschatological salvation, but it is not necessarily a prediction of individual healing miracles.

The importance of this text is not that it associates the coming of the Messiah with an eschatological action of God that includes the literal restoration of sight to the blind, healing, raising the dead, and preaching good news to the poor, but that it provides an example of this type of language being applied to the end time, even if only in a metaphorical sense. This provides a background against which Jesus’ reply to John the Baptist makes sense, not as a literal fulfilment, but as a novel interpretation of traditional expectations. We shall return to this point later.

Some Demonological and Apotropaic Texts

a. Introduction

In addition to the three texts discussed above, the material discovered at Qumran contains several references to evil spirits and to various forms of defence against them. Unfortunately, the poor state of preservation of many of the relevant texts often makes it hard to determine either the precise nature of the demonic affliction envisaged, or the precise means by which it is to be countered. Nonetheless, taken together these texts constitute important evidence for beliefs about demons that rarely surface in other Second Temple literature. In what follows, I shall first briefly set the scene from the Community Rule and the Damascus Document (two relatively unproblematic texts) and then go on to examine a number of the relevant demonological and apotropaic texts.

b. 1QS 3.13–4.1 and CD 12.2–3

At first sight the spiritual dualism described in the Community Rule looks more akin to the later rabbinic doctrine of good and evil inclinations, or even to the flesh and spirit in Paul, than to the practice of casting out demons. For the characteristic teaching of the Qumran sect is that the fate of human beings is determined by the influence that two spirits, the Prince of Light and the Angel of Darkness, have upon them (1QS 3.13–4.1). These spirits prompt individuals to good or bad behaviour; indeed the balance of spirits within individuals determines whether they are good or bad, sons of light or sons of darkness. But the doctrine says nothing about the evil spirits causing any other ill effects, such as possession or disease. The role of these spirits seems purely in the moral sphere. As the brief discussion of evil spirits in the Testaments of the Twelve Patriarchs showed, one strand of Enochic Judaism was already tending in this direction, although there the evil spirits appeared to be derived from those that afflict humankind not just with temptation but with disease and other misfortunes.

This more ‘popular’ view of evil spirits, as afflicters and not just tempters, certainly surfaces elsewhere in Qumran texts. One passage that is relevant in this regard is CD 12.2–3, which states that ‘Every man who preaches apostasy under the dominion of the spirits of Belial shall be judged according to the law relating to those possessed by a ghost or familiar spirit (Lev. 20:27).’ Vermes takes this to be a reference to ‘apostasy in a state of demonic possession’.67 If this interpretation is correct, it would surely indicate that demonic possession was known to the writers of the Damascus Document. Indeed, the allusion to the biblical text at CD 12.3 is by means of the phrase האוב והידצוני, referring to the two kinds of spirit condemned at Lev. 20:27 (אוב או ידצני). These are terms that, in the Hebrew Scriptures, appear to relate to illicit forms of spirit possession.

One might argue that the reference to the ‘spirits of Belial’ at CD 12.2–3 is simply polemical. Whereas the Damascus Document does not explicitly teach the two-spirit doctrine of the Community Rule, it does polemicize against its opponents by representing them as being under the dominion of Satan, for example at CD 5.11–19. This passage concludes by likening the sectarians’ present-day opponents to the magicians Jannes and Jambres. Someone capable of writing this is quite capable of regarding a preacher of apostasy as being demonically inspired ipso facto. That said, the possibility of condemning someone as demonically inspired presupposes a general belief in the possibility of demonic inspiration.

c. 4Q510 and 4Q511

4Q510 and 4Q511, ‘Songs of the Sage (or Maskil)’, also attests a belief in the demonic. At first sight 4Q510 reads as a psalm of praise and thanksgiving to the God who is able to put the forces of darkness to flight. But the purpose of this appears to be to frighten the evil spirits into leaving the community well alone:

And I, the Master, proclaim the majesty of his beauty to frighten and ter[rify] all the spirits of the destroying angels and the spirits of the bastards, the demons, Lilith, the howlers (?) and [the yelpers …] they who strike suddenly to lead astray the spirit of understanding and to appal their heart and their … in the age of domination of wickedness and the appointed times for the humiliation of the sons of ligh[t], in the guilt of the ages of those smitten by iniquity, not for eternal destruction but for the humiliation of sin.

This text follows a fairly common apotropaic pattern in which appeal is made to some attribute of God, such as his majesty, in order to counter evil spirits. This purpose of these texts is also indicated by a fragment (fr. 8, 1.4) of 4Q511:

[For the Master]. Second [S]ong to frighten those who terrify him …

These ‘Songs of the Sage/Maskil’ raise a number of questions, however. The first concerns the nature of the spirits that are to be frightened and terrified. The reference to the ‘spirits of the bastards’, both in the passage cited above and in a number of the fragments of 4Q511 (e.g. fragment 35), suggests a close connexion with the aetiology of demons encountered in the Book of Watchers and Jubilees, since the ‘bastards’ are presumably the offspring of the union of the fallen angels with the daughters of men. Moreover, the reference to God’s sealing of the abyss in 4Q511 fragment 30 could be an allusion to the imprisonment of the Watchers, which would further strengthen the connexion to the Enochic mythology. This, together with the designation ‘the demons, Lilith, the howlers’ most naturally suggests that the popular or individual-afflicting conception of the demonic is in view. On the other hand, these evil spirits are then described as ‘they who strike suddenly to lead astray the spirit of understanding and appal their heart’. This suggests rather that the demons are acting in the moral sphere, as agents of the Angel of Darkness trying to draw the sons of light off the path of righteousness. Leading people astray was already one the functions of the demons in Jubilees who attacked Noah’s descendants after the flood, and there it was combined with the ‘afflicting’ role of demons in causing illness and death. It is not so clear that this is the case here. The ‘Songs of the Sage/Maskil’ appear to be a sectarian composition, and it is thus reasonable to interpret their demonology against the background of the sectarian two-spirit doctrine. Against this background it is at least possible that evil spirits would be blamed more for temptation and sin than for illness and possession. The fragmentary nature of these texts makes it impossible to be sure, and makes any argument from silence doubly perilous, but the absence of any explicit reference to any but moral, emotional and intellectual attacks should make one hesitate to describe their activity either as possession or affliction in the sense of causing disease.77

It should also suggest caution in describing the Maskil’s apotropaic counter-measures as ‘exorcism’. For one thing, the word ‘exorcism’ should be reserved for the cure of demonic attack, whereas the aim of the Maskil’s prayer seems rather to be prevention. For another, ‘exorcism’ in the strict sense of driving a demon out of a possessed person, or even in the slightly looser sense of driving a demon away from an afflicted person, hardly seems to be in view here. As P.S. Alexander aptly remarks:

The scenario envisaged is not an exorcism, at least if we take exorcism in its precise sense of expelling a demon from the body of an individual. Rather the Maskil is, through prayer, erecting or maintaining a spiritual cordon round the Community, pre-emptively to keep at bay the forces of darkness.

Furthermore, although the Maskil’s prayer here is frequently classified as ‘magical’, it is questionable whether it is so in the sense of the definition of ‘magic’ given in Chapter 1 and defended in Chapter 13 below. The Maskil frightens the spirits by proclaiming ‘the majesty of his [sc. God’s] beauty’, having sung ‘Ben[edictions for the K]ing of glory. Words of thanksgiving in psalms of … to the God of knowledge, the Splendour of power, the God of gods, Lord of all the holy’ whose ‘domin[ion] is over all the powerful mighty ones’. In other words, God is clearly the (potential) BNP here, with the Maskil perhaps acting as the (potential) PNP, although in fact the surviving fragments of 4Q510 and 4Q511 contain no indication that the Maskil directly attempts to prompt God to act. Rather, as Alexander aptly expresses it, ‘Basically the Maskil warns the demons not to meddle with him and his Community, because they have got “protection”.’ Appealing to God’s protection is, in itself, hardly magical.83

d. 4Q560

Another Qumran text intended as a defence against demons, 4Q560, is more clearly magical:

Fragment 1 Column 1 (1) … heart … (2) a new mother, the punishment of those giving birth, a command (of) evil … (3) the male poisoning-demon and the female poisoning-demon (is forbidden) [to] enter the body … (4) [(I adjure you) by the Name of He who forgiv]es sins and transgression, O fever and heartburn (5) [… and forbidden to disturb by night in dreams or by day] in sleep, the male PRK-demon and the female PRK-demon, those who breach (?) …

Fragment 1 Column 2 (2) … before h[im] (4) before him and … (5) and I adjure you, O spirit … (6) I adjure you, O spirit … (7) upon the earth, in the clouds …

This appears to be a formula for use in an amulet to protect against demonic attack, although the physical form of the fragment makes it unlikely that this copy was actually used in an amulet. Michael Wise suggests that it probably comes from a magical recipe book, and remarks, ‘Particularly striking is the mention of the Fever-demon, for this same demon appears in the New Testament’ (referring to Lk. 4:38–39). His claim that ἐπετίμησεν τῷ πυρετῷ at Lk. 4:39 can be translated ‘rebuked the Fever-demon’ perhaps goes too far, although this may perhaps be an acceptable paraphrase of what Luke intended, in which case the parallel would be pertinent. The assertion by Eisenman and Wise that ‘Exorcisms and conjuring of this kind were also very popular activities in the Gospels and Acts’ should also be treated with caution. In the New Testament only the Jewish exorcists at Acts 19:13 attempt to exorcise an evil spirit using the formula ‘I adjure you by …’; in the Gospels this formula is more noticeable by its absence from the lips of Jesus.

What makes this fragment of particular interest is its close resemblance to other magical formulae. The detailed analysis by Douglas Penney and Michael Wise identifies five features of this text that relate it closely to Jewish magic texts previously known only from manuscripts several centuries later. For example the Aramaic expression used in line 3 for ‘to enter the flesh’ (עלל בבשרא) uses the verb and preposition that are normally employed together to express demonic possession, although incantation texts more usually speak of the body (פגרא) rather than the flesh (בשרא). The pattern ‘X male-demon and Y female-demon’ (lines 3 and 5), the appeal to some attribute of the countering deity (line 4) and the formula ‘I adjure you …’ are also characteristic of magical texts from a wide range of Middle Eastern sources.

The demons envisaged by this text do appear to be more of the individual/afflicting type, since ‘It contains an adjuration against demons which attack pregnant women, cause various illnesses and disturb sleep.’ There is nothing in the text to indicate that it is of sectarian composition,93 as indeed one might expect on the basis of its similarity to much later magical texts. The presence of this fragment at Qumran thus provides important evidence that the more popular, individual-afflicting conception of evil spirits existed alongside the doctrine of their morally corrupting role. Such a combination is not particularly surprising, but it may be relevant to the interpretation of other Qumran texts where it is less clear what form of demonic attack is being repulsed.

In what sense 4Q560 actually concerns exorcism is not clear, since the provision of a general purpose formula for protection against demonic attack is not the same thing as expelling a demon that has already attacked a particular individual. Concerning this text Alexander justly remarks, ‘The spell which it contains is a general apotropaic. It does not imply that the wearer is actually afflicted with the illnesses mentioned (he or she would be in a sorry state if they were). Rather it would be worn to prevent such illnesses striking.’ Once again we are in the realm of prevention rather than cure.

e. 11Q11 (11QapPsa)

The scroll 11Q11 (=11QapPsa) may contain the text of prayers or psalms intended to expel a demon from one afflicted, which could be evidence of exorcism in the strict sense. For example, in Vermes’ translation part of Column 4 reads:

To David. O[n words of incanta]tion. [Cry out al]l the time in the name of the Lor[d]

towards heave[n when] Beli[al] comes to you [And sa]y to him:

Who are you? [Be afraid of] man and of the seed of the ho[ly ones].

Your face is a face of [nothin]g and your horns are horns of dr[ream].

[You ar]e [d]arkness and not light; [injustice and not righteousness.

[The prin]ce of the h[os]t [is against you]; the Lord [will cast] you [to] the nethermost hell

The number of (conjecturally completed) lacunae in this text is one indicator of the difficulty of interpretation. For example, in the translation of the same passage offered by Garcia Martínez and Tigchelaar, the lacuna in the fourth line that Vermes completes with ‘Be afraid of’ is instead reconstructed as ‘oh offspring of’, so that the line then reads, ‘Who are you, [oh offspring of] man and of the seed of the ho[l]y ones?’ This creates a clear reference to the Enochic myth of the fall of the Watchers which is absent from Vermes’s version, and which completely alters the way in which the phrase ‘holy ones’ is to be understood. Edward Cook offers yet another way of completing this lacuna when he translates this same line ‘Who are you? [Withdraw from] humanity and from the ho[ly] race!’98 The point is not to criticise any of these reconstructions, but to urge due caution in building interpretations on such an uncertain basis.

The basis for supposing that this text concerns exorcism is partly that threats of divine and angelic action are to be made against Belial when he ‘comes to you’, that is when some form of demonic attack is already in progress. It is also partly that column 5 of this fragment continues with a version of Psalm 91, which subsequent Jewish tradition recommends for use against demonic attack. To this must be added the many references to demons throughout the text (especially in column 1) and to the threats of divine retribution against the person addressed (most probably a demon) in column 3. Read against the mythology of the Book of Watchers, the language of column 3, which talks about a mighty angel being sent to despatch the addressee into the great abyss, could theoretically be applied to the Watchers rather than their demonic offspring, but this is unlikely in practice. Since the Watchers have already been bound pending the final judgment, there would be no need to threaten them with further punishment. Instead, this is probably another example of the tendency to assimilate language about demons to language about the Watchers.

Granted that the psalms in this text are intended in some way to ward off demonic attack, there still remains the question of what form the attack is taking. The apocryphal psalms of the first four columns give very little clue, since in the text that survives they are mainly concerned with uttering threats based on God’s power, wrath and intention to punish some unspecified party in the abyss. It is true that 4.12 perhaps talks of a demon doing harm to a righteous man, which might suggest that the demon is inflicting an illness. But this assumes that the single ש at the end of the line is to be taken as the first letter of שד (demon) rather than any other Hebrew word beginning with ש. It further assumes that the preceding words הרע לו are to be vocalized to give the sense ‘evil to him’ (perhaps הָרַע לוֹ), though given the fragmentary state of this line it is at least conceivable that רע represents not רַע (‘evil’) but רֵעַ (‘friend’, or even ‘thought’ or ‘noise’). Again, the point is not to criticise the reconstructions ventured by various translators, but to draw attention to the danger of a circularity in which a notion of what the text is about is used to reconstruct the text, and the reconstructed text is then employed to support an argument for what the text is about.

Alexander tries to avoid this trap by appealing to three features of the extant text that indicate that individual illness is in view: first that the person exhorted to use the psalms is addressed as ‘you’ singular, secondly the reference to plague and pestilence in the citation of Psalm 91, and thirdly the references to healing in the text of the apocryphal psalms. The first of these points seems to be correct. The second is less clear. Psalm 91 certainly does contain language that would be consistent with desiring protection from illness, but it is possible that it could also be used more generally. The third argument relies on the presence of רפואה at the end of the poorly preserved line 7 of column 2 and [ר]פאל שלמ[ם] in the equally poorly preserved line 3 of column 4, which is then translated ‘Raphael healed’. To be sure, Raphael’s name means ‘God has healed’, but given that this angel’s role in the Enochic myth is to punish the Watchers, it is at least possible that מלש should be taken as (part of) the piel form ‘requite’. Again, given the possibility that the word ‘heal’ can take on a metaphorical sense of spiritual healing, that is curing sin, it is also conceivable that רפואה at 4.3 bears this non-literal sense. Against the possibility that this text envisages demons inflicting individuals with illness must be weighed the other possibility that it envisages them tempting them or misleading them into sin, and that the function of the psalms is to ward off what we might call this ‘moral’ form of attack. It would be far from inappropriate to address the words ‘You are darkness and not light; injustice and not righteousness’ to an agent of the Angel of Darkness who is trying to lead one astray.

To decide between these alternatives it would be helpful to know whether these songs were a sectarian composition (or compilation), and hence quite possibly designed for use primarily within the particular framework of Qumran dualism rather than the wider framework of popular demonological beliefs. That 11Q11 was written out at Qumran seems likely, since the handwriting appears to date from the turn of the era. This, however, hardly settles the question of the text’s original composition. In Lange’s view, the text cannot be of sectarian origin, since it makes such free use of the tetragrammaton.105 Alexander, on the other hand, does not find this argument convincing since ‘There are other probably sectarian scrolls in which the Tetragram is written fully and without substitutes’. In Alexander’s view the text probably is sectarian, though it contains nothing either obviously non-Qumranic or characteristically Qumranic, and may well reflect more widely held views. At this point it unfortunately becomes apparent that to appeal to the supposed origin of the text to support one line of interpretation is likely to result in a circular argument.107

That said, the balance of probability would seem to favour an individual-afflicting view of the demonic in this text. At the very least, there is little or nothing that is strictly incompatible with the view that the demons to be warded off are those that cause illness. This is certainly a plausible reading of Psalm 91 in this context, and this reading is lent considerable further support if 11Q5 (on which see below) can be taken as referring to the four psalms in the present text. But it must be stressed that such a balance of probability makes this reading of the text only the more likely of two perfectly possible hypotheses (the other being that the psalms are to be recited by persons concerned about demonic temptations), and therefore that probability should not be allowed to harden into certainty. Any hypothesis about the exorcistic nature of this text thus rests on relatively insecure foundations.

It is certainly possible to make out a case the other way. If Vermes is correct in conjecturing ‘Belial’ in his translation of 4.5 this would be problematic for the sickness theory. For as Alexander argues, Belial is too important to be a mere demon, since ‘Demons are the foot-soldiers of the forces of darkness’ whereas Belial ‘is obviously a very powerful figure, the chief opponent of Michael and the good angels’. Belial ‘must, therefore, belong to the category of angel’ but ‘If he is an angel, then he cannot be the direct cause of the illnesses envisaged in 11Q11, since an angel cannot penetrate the body of a human. That is possible only for a demon or an evil spirit.’ Vermes’ conjectural ‘Bel[ial]’ is supported by the surviving letters בלי in the text, and is thus highly plausible (though בלילה, ‘by night’, is another possibility here). If it is also correct then the words that follow in Column 4 are to be addressed to Belial and so, on Alexander’s own argument, cannot be intended to exorcise a demon that is causing illness or possession (though they could be intended to ward off temptation).

Even allowing that these apocryphal psalms may be exorcistic in function, it needs to be determined in what sense they are so. First, there is nothing in 11Q11 that indicates ‘possession’ in the narrow sense. In fact, ‘possession’ in the narrow sense of an evil spirit taking control of the sufferer’s personality seems to be excluded by the suggestion that these are psalms to be recited by the sufferer. This does not, however, exclude a form of demonic attack in which the evil spirit has invaded the sufferer’s body to cause illness, and thus needs to be driven out in order to effect a cure. Secondly, the tendency to regard these psalms as magical incantations needs to be questioned; at the very least one would not wished to become entrapped in a circular argument in which the psalms are deemed to be ‘incantations’ because they are clearly magical, and are regarded as magical because they have been labelled ‘incantations’, the underlying assumption being that any type of exorcism is a form of ‘magic’. Whether or not exorcism was a form of magic was probably a question on which Second Temple Jews were sharply divided. In other contexts, one would not think of calling the recitation of Psalm 91 an ‘incantation’.

Alexander avoids this kind of circularity by finding a ‘technical description of the text … as “words of incantation” ‘ in the ‘opening rubrics’ of the psalms. But a certain amount of speculation is once again needed here. For one thing, Alexander bases his argument on the superscription at 4.4 and then asserts of this and the previous psalm, ‘Their opening rubrics were probably the same: לדויד על דברי לחש בשם יהוה.’ Unfortunately the key phrase דברי לחש, rendered as ‘words of incantation’, is represented in the surviving text only by the final two letters חש, the remainder being a conjectural completion of a lacuna that begins after ע לדויד. Admittedly ‘incantation’ is a highly probable conjecture, since there are not that many Hebrew words ending in חש and scarcely any of the alternatives would fit the context so well. But whether this is sufficient evidence to show that the rubric intends to denote the psalm as an incantation in some technical sense remains open to question. Perhaps a more telling formal feature of these poems is the conclusion ‘Amen Amen Selah’, which is characteristic of Jewish magical incantations rather than biblical poetry. Unfortunately, in the three places where this formula appears in the text (4.3; 5.3, 14), the ‘Amen Amen’ results from a conjectural completion of a lacuna. Now, whereas the phrase ‘Amen Amen Selah’ is not typical of biblical poetry, ‘Selah’ by itself is often found in the canonical Psalms, and is in fact found by itself in this manner at 11Q11 5.6 (citing Ps. 91:4, although Ps. 91:4 lacks ‘Selah’). Thus, whereas ‘Amen Amen Selah’ may represent a plausible completion of the lacuna, to the extent that this conjectural completion is based on parallels from Jewish magical incantations any argument that this formula shows 11Q11 to be such an incantation is clearly circular.

It is also open to question in what sense this kind of procedure should be described as ‘magical’ (as, for example, both Lange and Alexander take it to be). To be sure, this must depend on how one chooses to use the term ‘magic’. On the definition to be defended in Chapter 13 below, something would be counted as magic if the source of superhuman power needed to perform it were other than God. The psalms in 11Q11 on the other hand appear to be appealing to God’s power in one way or another. If they were to be regarded as calling upon God to act, this might reasonably be called ‘prayer’ rather than ‘magic’. If they were being used to coerce evil spirits to perform some service then this would be magical. Their use to frighten spirits off by appeal to God’s power appears to lie somewhere between these two, but it is not too difficult to imagine how devout people who accepted the Deuteronomistic prohibitions against magic could regard self-defence against demons to be perfectly legitimate, and hence not magical, especially if the principle means of that self-defence was an appeal to the power of God in whom they had sincere trust. Whether or not the people who used this texts did regard them in such a way, or whether they simply employed them as an apotropaic formula effective in itself is something the text cannot tell us; but perhaps this is in any case not a distinction the people at Qumran would have made.

To summarize: 11Q11 appears to contain a collection of perhaps four psalms, including a version of the canonical Psalm 91, intended for use against evil spirits. But the poorly preserved state of the text makes it hard to be either certain or precise about their function. In particular, it is not entirely clear whether the evils spirits to be opposed are conceived primarily as attempting to lead their victims away from the path of righteousness or as afflicting them with disease. It is also not clear whether the apotropaic use of these psalms should be regarded as prayerful, magical, or perhaps something in between. Comparison with other texts perhaps tips the balance of probability in favour of disease-causing demons being countered by quasi-magical incantations, but this remains an uncertain hypothesis. One Qumran text that lends particular support to this hypothesis is 11Q5, which will therefore be discussed next.

f. 11Q5 (11QPsa)

The scroll 11Q5 (=11QPsa) contains a number of canonical and non-canonical psalms. Among the latter appears the following notice at column 27:

David Son of Jesse was wise and brilliant like the light of the sun; … YHWH gave him an intelligent and brilliant spirit, and he wrote 3,600 psalms and … 4 songs to make music on behalf of those stricken (by evil spirits) … (27.2, 3, 4, 10).

Notwithstanding the reference to Solomon at 11Q11 1.2, the psalms of 11Q11 appear to be attributed to David and, as we have seen, appear also to be intended for use against affliction by evil spirits. It is thus plausible to suppose that the four psalms of 11Q11 are the four referred to here. The reference in 11Q5 to David composing songs for those afflicted presumably reflects the story in 1 Sam. 16:14–23, which has already been mentioned in connexion with Josephus and LAB 60. The evil spirit that troubles King Saul in these texts torments him, and is thus to be understood as belonging to the individual-afflicting type. This tends to suggests that if 11Q11 contains the apotropaic psalms supposedly composed by David, their function was to counter this type of evil spirit rather than to ward off temptation. Furthermore, there is a suggestion that David’s psalms are effective by virtue of the brilliance of their composition, and thus belong in the same category as the incantations against demons composed by Solomon according to Ant. 8.45–49. The procedure described by Josephus certainly looks magical, and this suggests that the ‘Davidic’ compositions of 11Q11 should be seen in the same way.

This conclusion can only be tentative, since it is based upon a number of suppositions that can only be more or less probable. In particular, the fact that Josephus describes an exorcism employing Solomonic techniques in a way that makes it look distinctly magical does not prove that all other Jews understood exorcism in the same way; it is simply a suggestive parallel.

g. 4Q266, 4Q269, 4Q272 and 4Q273 (on Skin Disease)

There is at least one further text that could suggest that the Qumran sectarians believed that demons could afflict people with disease as well as leading them into error. The fragments of a version of the Damascus Document found in 4Q266, 4Q272, 4Q273 and 4Q269 describe the cause of a skin complaint in terms of the invasion of a spirit, but it is not clear whether the spirit in question is a demonic one, or whether the term רוח is being used in a purely physiological sense here. Armin Lange takes this text to refer to a ‘magical view of ringworm’ in which ‘ringworm at head or beard is caused by a spirit which enters the body’. But this is not the only possible view. For example Joseph Baumgarten states:

In the light of the ‘two-spirit’ theology set forth in the catechism of the Community Rule (1QS 3–4), one might readily attribute the genesis of disease to the spirit of evil … It is thus possible to take the attribution of scale disease to the רוח in our text as involving the intrusion of evil or demonic influences.

But he at once continues, ‘However, one cannot fail to note that the רוח is mentioned here in the context of a very concrete physiological condition.’ He goes on to observe that other Jewish medical texts seem to have a theory about blood-flow and skin disease involving רוח in much the same way as πνεῦμα features in the Greek theory of the humours. He further observes:

This identification of the blood with the spirit of life would seem to be more in harmony with the biblical perspective (Gen. 9:3–4 and Lev 17:10–14). It is also characteristic of the early Jewish medical compendium, the Sepher ha-Rephu ʾat, attributed to Asaph.

One could thus make a case for denying any reference to evil spirits here; but even if the text does in fact envisage a form of possession in which an invading demon causes skin disease, the instructions given appear to be purely diagnostic, so that it cannot be said that an exorcism is directly in view.


The fragmentary nature of the texts examined in this chapter is frustrating, since in many places where they look as if they might provide fruitful parallels to the healing and exorcism stories of the Gospels, one is forced to resort to a considerable measure of conjecture in their reconstruction and interpretation. In particular, it is not always clear how far references to disease and healing relate to the cure of physical ailments or (say) of sin, or how far evil spirits are seen as afflicting the individual sufferer with disease or possession, or as serving as part of Belial’s campaign to mislead the sons of light. To be sure, the alternatives posed here are by no means mutually exclusive, and taken in aggregate, the Qumran texts examined here do seem to show some concern with physical illness and demonic affliction. There are two reasons in particular why this is significant for the present enquiry.

The first is that the rigid ‘two-spirit’ doctrine as it is found in the Qumran sectarian literature may well be peculiar to the sect. This ‘official’ sectarian teaching would presumably wish to maintain that the most harmful thing Belial or his minions could achieve would be to befuddle a child of light into becoming a child of darkness (which might be seen as danger to be guarded against despite its tension with the official line on determinism). Given that, indications of evils spirits as also afflicting people with disease (and perhaps possession in the strict sense) presumably reflect popular beliefs that are more widespread.

The second is that the Qumran texts represent a selection of writings that happen to have survived the vagaries of two millennia, rather than those that were deemed worthy of preservation and copying by some particular group. This allows the possibility that these texts could preserve popular beliefs from around the turn of the era that other lines of transmission have neglected or edited out. Taken together, these two factors could indicate that the views on evil spirits and counter-measures against them expressed in the Qumran texts were far more widespread in first-century Judaism than one might suppose from reading Philo, Josephus and a large number of the Apocrypha and Pseudepigrapha.

That said, some caution is also in order. Even if the Qumran texts reflect more widespread popular beliefs, these beliefs may have been given a distinctively Qumranic flavouring in the community. Moreover, the parallels with the Gospel healing and exorcism stories should not be made stronger than they are. None of the Qumran texts examined unambiguously envisages demonic possession in the strict sense in which it appears in the Synoptic Gospels. With the exception of the Genesis Apocryphon, none of these texts gives an account of an exorcist or healer at work, and although the account of Abraham’s dealings with the king of Egypt in the Genesis Apocryphon does bear some resemblance to the Gospel healing and exorcism stories, there are also important differences, which we have noted. The other relevant texts that appear to describe exorcistic or healing techniques do so in terms that are notably different from the methods described in the Gospel accounts of Jesus. In particular, Jesus is never described as employing incantations composed by some great sage of the past such as David or Solomon. To try to represent Jesus’ brief commands to demons as magical incantations on the basis of this or that formal parallel to the material examined in this chapter would be to force the evidence.

In general terms one might say that the material reviewed in this chapter exhibits a context of belief in demons against which Jesus’ exorcistic activity makes general sense. There are, however, three more specific points of contact. The first is that the Qumran texts show an approach to the demonic within a general Enochic framework. This allows evil spirits to be seen both as capricious demons maliciously afflicting individuals with various kinds of suffering and as agents of a more universal evil spiritual power systematically opposed to God. In short, it allows the unity of evil presupposed by Jesus’ arguments in the Beelzebul controversy (Mk 3:22–30; Mt. 12:22–32)—but not necessarily accepted by his opponents. The second is that the Qumran texts virtually all presuppose in some way or another that it is God’s power that is needed to repel demons. In 11Q11 it seems that a troublesome evil spirit is to be threatened with the power of the God who imprisoned the Watchers. This kind of belief provides a plausible context for the otherwise puzzling saying at Mt. 12:28, puzzling because it is otherwise not clear why exorcisms should be seen as having any eschatological significance. But just as Jesus may have made a novel interpretation of the kind of beliefs expressed in 4Q521 to give eschatological significance to his healings, so he may have made a similarly novel interpretation of the Enochic myths so that his exorcisms became signs of the coming of God’s rule. The third is the ambiguity surrounding the nature of the Qumran material: magic, miracle or just prayer? It is entirely plausible that the people who used apotropaic texts at Qumran thought of themselves as legitimately appealing to the protection of the all-powerful God in whom they trusted while their enemies could have stigmatized their procedure as ‘magical’ on the basis that they were employing ‘incantations’ in direct defiance of the Deuteronomic prohibitions. The Beelzebul controversy indicates that the same kind of ambiguity surrounded Jesus’ exorcisms.

These Qumran texts have taken this survey a long way from its starting point in the writings of Josephus and Philo, in terms both of the ideas expressed and of the manner of treatment. The final two texts to be examined, Tobit and Artapanus, will return closer to the starting point. Both are narrative texts, and in Artapanus we shall meet yet another retelling of the career of Moses.

Chapter 8

Miracle and Romance


The previous two chapters have, inter alia, traced some of the Enochic traditions about fallen angels through the Book of Watchers and Jubilees to their influence on some of the demonological and apotropaic texts discovered at Qumran. These traditions find a further echo in the book of Tobit, in which an angel called Raphael once again binds an evil spiritual being with a taste for human females. Tobit also continues the demonological theme by containing a story of an exorcism.

The use of the heading ‘Miracle and Romance’ to link Tobit with the second text to be discussed in this chapter, Artapanus, is admittedly artificial. To be sure, both Tobit and Artapanus might loosely be termed ‘romances’, but whereas Tobit also has strong affinities with wisdom and paraenetic literature, Artapanus veers more towards historiography, or at least legendary historiography for polemical or propagandist purposes. Both in tone and content it is a very different work from Tobit. The two works seem also to have stemmed from very different circles. Whereas Artapanus is almost certainly the product of Egyptian Judaism, in Tobit Egypt features only as the distant land to which Asmodeus flees, perhaps suggesting a Galilean or eastern diaspora provenance for the work. Again, while there is little reason to doubt that Artapanus originally composed his work in Greek, the original language of Tobit seems to have been Aramaic (see below).

Yet there are, perhaps, two reasons why it might be appropriate to treat these two texts in the same chapter, beyond the fact that they are both prose narratives containing miracle stories. The first is that both texts may well have been written around the same time, say the third or second century bce. The second is that both texts, albeit in quite different ways, provide potentially interesting parallels to the Gospel accounts of Jesus. In the case of Tobit this is constituted by the presence of both a healing story and an exorcism. In the case of Artapanus, it is constituted rather by the way in which the text tends to present its central character as an immanent BNP. We shall now consider each text in turn.


As we have seen, actual stories of healing miracles and exorcisms are comparatively rare in Second Temple literature. The Book of Tobit appears to furnish an example of each. The eponymous hero is blinded by sparrows defecating in his eyes (Tob. 2:10); he is subsequently cured through the good offices of the angel Raphael; the angel commands Tobit’s son Tobias to anoint Tobit’s eyes with the gall of a fish, ‘and when they smart he will rub them, and will cause the white films to fall away, and he will see you’ (11:8 rsv). This is seemingly one of the closest parallels in the literature under examination to the healing stories of the New Testament. With the film that fell away from Tobit’s eyes one might compare the ‘something like scales’ that fell from Paul’s (Acts 9:18), and with the material means (fish gall) employed by Tobias one might likewise compare the material means employed by Jesus upon the blind men at Mk 8:22–26 and Jn 9:1–7. Furthermore, on the face of it, Tobit provides one of the closest analogies in Second Temple literature to the Gospel exorcism stories. The pious Jew Sarah is troubled by a demon, Asmodeus, who kills each of her seven successive husbands on their wedding-night. Tobias, however, has been instructed by Raphael to deal with the demon by smoking the gall and liver of a fish. So when he marries her he carries out these instructions, with the result that Asmodeus flees to ‘the remotest part of Egypt’ where Raphael binds him (Tob. 8:3 rsv).

But before one can usefully compare either of these incidents with the Gospel accounts of Jesus, it is necessary to enquire into the nature of this material. For one thing, the book of Tobit exists in several recensions, and one needs to discriminate between them. For another, it makes use of folkloric and possibly magical traditions that may not be central to its Jewish author’s main purpose.

Following David Simpson, I shall refer to the three principal Greek recensions of Tobit as Rv, Rs and Rc. Rv, the text used as the basis of the rsv translation, is a shorter version of the book based on Codices Vaticanus and Alexandrinus and supported by a large number of miniscules. The longer recension Rs is based primarily on Codex Sinaiticus, supported by the Old Latin version, which, together with a couple of miniscules (319 and 910) can be used to supply the omissions in the Greek text of Sinaiticus. Rc appears to be a later mixed text found in a few miniscules, and extending only over Tob. 6:9–13:8. Fragments of Tobit in Hebrew and Aramaic have also been found at Qumran, and in the main support the readings of Rs. If this suggests that Rs is likely to be the earliest form of the Greek text, this has also been demonstrated on other grounds, with arguments suggesting that Rv is a later redaction of Rs. It is also generally believed that Rs is a translation from a Semitic original, probably in Aramaic. For the purposes of the present discussion, therefore, I shall take Rs to be the basic text, although I shall note the variant readings of Rv where these are pertinent to the argument.

All the complete recensions of Tobit tell the same basic story. The eponymous hero, formerly a native of the northern kingdom of Israel, has been taken into exile in Nineveh. A devout man, he is particularly punctilious about giving alms and burying dead Israelites. This latter activity gets him into trouble with the Assyrian authorities, but he persists in burying yet another of his fellow countryman when his son Tobias, sent by his father to find a poor Israelite to join their family in a feast, instead finds a dead one. Resting from his labours, Tobit falls asleep outdoors and is blinded by sparrow-droppings. For two years his relative Ahikar supports him, but when Ahikar moves away, Tobit is forced to depend on the earnings of his wife Anna. When Anna one day returns with a bonus from her employer in the form of a young goat, Tobit accuses her of stealing it. He is so depressed by the ensuing row with his wife that he prays for death. At the same moment a female relative in distant Ecbatana is also praying for death, having been bitterly reproached by a servant girl for killing off her husbands—seven of them in succession (although the reader is informed that is actually the evil demon Asmodeus who has killed them). Their stories become linked when Tobit despatches his son Tobias to fetch money previously deposited with another kinsman at Rages. The angel Raphael, disguised as an Israelite called Azariah, acts as Tobias’s guide, and advises him how the innards of a fish caught along the way can be used to cure blindness and drive away demons. At the same time, Raphael tells Tobias about Sarah and encourages him to marry her, since she is his relative and she has been destined for him in heaven. Raphael overcomes Tobias’s fear of marrying a woman with a demonic guardian who kills off her husbands by instructing him how to fumigate the demon, and Tobias immediately falls in love with her before even meeting her. The angel and the lad thus stop off at Ecbatana, and Tobias requests Sarah’s hand in marriage, despite her father’s understandable reservations. Her father, Raguel, even goes so far as to dig a grave for Tobias when the latter goes in to his daughter on their wedding night. But this proves unnecessary, for Tobias follows Raphael’s instructions and drives the demon away by burning the heart and liver of the fish. Raphael collects the money deposited at Rages, then returns home with Tobias, Sarah and a large dowry from Raguel. On arriving home, Tobias applies the gall of the fish to his father’s eyes, and Tobit’s sight is duly restored (along with his economic fortunes). When Tobit and Tobias try to pay ‘Azariah’ for his services, the angel reveals his true identity and ascends back to heaven. After Tobit and Anna die (at a good old age), Tobias and Sarah move back to Ecbatana, and Tobias lives to see Nineveh destroyed.

Several scholars have suggested that a number of secular folkloric motifs lie behind this tale, including the Tale of Ahikar, the Grateful Dead Man, the Tale of Two Brothers, and the Monster in the Bridal Chamber. Although the dependence on the Ahikar story is generally acknowledged, that on some of the other alleged sources has been questioned.9 Nevertheless, it seems clear that one or more folk tales have been employed in the composition of Tobit, even if it cannot now be identified precisely what they were. If it is these secular folk tales that have supplied most of the details for the healing and the exorcism, one will have to question how far they really represent Jewish ideas of miracle.

On the other hand, any folkloric material employed by Tobit has been placed in a thoroughly biblical context, so that the book’s dependence on the Hebrew Bible is unmistakable. For example, the finding of a suitable endogamous bride for Tobias is strongly reminiscent of the finding of a bride for Isaac in Genesis 24, and the fortunes of Tobit in Nineveh bear some resemblance to those of Joseph in Egypt. There are also some correspondences between the Sarah who married Tobias and the Sarah who married Abraham. The similarity between their two stories becomes even more striking in the Genesis Apocryphon, in which the king of Egypt is afflicted by a demon who prevents his consummating his forced ‘marriage’ to the biblical Sarah. Above all, Tobit is heavily dependent on Deuteronomic theology. In particular, despite their initial reverses the main characters illustrate the Deuteronomic theory of rewards and punishments, and do so in a manner that seems designed to promote a similar style of piety among Diaspora Jews living under foreign rule. Moreover, the healing of Tobit and Sarah is placed within the wider context of the exile in such a way as to foreshadow the salvation, not simply of pious individuals, but of Israel as a whole.15 Given that the folkloric material in Tobit is made to serve these wider concerns, one wonders how far the author was really interested in the healing and the exorcism for their own sake. Any suspense in these matters is removed by the narrator’s notice at 3:16–17 that God sent Raphael to heal Tobit and Sarah and to bring about the latter’s marriage to Tobias. Given the threat that Asmodeus is meant to present to Sarah’s husbands, the description of his fumigation by Tobias and binding by Raphael is so cursory as to justify Carey Moore’s regarding it as ‘anticlimactic’. Moore seems equally correct in remarking that ‘Although chapter 11 describes the recovery of Tobit’s sight, that is not what really interests either the narrator or the reader.’17

Given this relative lack of emphasis on the healing and exorcism and Tobit, one may well question Bernd Kollman’s suggestion that they serve to legitimate medical and magical means (at a time when these were suspect in Israel) by having them revealed through a divinely appointed mediator. Another feature of the text that makes this unlikely is the apparent polemic against the medical profession at 2:10b, ‘Although I went to physicians for healing, the more they treated me with various remedies, the worse my eyesight became until I lost my sight’ (cf. Mk 5:26). It might be argued that since, unlike Tobias, these bumbling physicians were not acting under divine guidance they do not undermine Kollman’s point. But if the author of Tobit were concerned to legitimate medical and magical means of healing, introducing such devastatingly incompetent physicians near the start of his tale would surely be counter-productive. Kollman nevertheless performs a valuable service by cataloguing ancient medical and magical parallels to the procedures used to deal with Tobit’s blindness and Sarah’s demonic persecution. In doing so, he casts doubt on how far these incidents can be regarded as strictly miraculous.20

If they are miraculous, it is hardly by virtue of the fact that Tobias performs an exorcism and a healing, assuming that both take place more or less by naturalistic means. In these acts Tobias is not obviously portrayed either as BNP, MNP or PNP (the petitioners were Tobit and Sarah, but neither of them directly requested healing), though this observation will need to be qualified in connexion with the exorcism. The involvement of Raphael in the binding of Asmodeus certainly complicates the issue, and this too must be discussed when we look at the exorcism in more detail. Raphael (whose name means ‘God heals’ or ‘God has healed’) perhaps represents the presence in the world of an otherwise transcendent God, but apart from binding Asmodeus, his only role in either the healing or the exorcism is in instructing Tobias what to do. To be sure, Raphael has been despatched by God for this very purpose in response to the prayers of Tobit and Sarah, and this does place both the healing and the exorcism under the umbrella of divine action. Furthermore, it is surely no coincidence that a suitable fish appears for Tobit to catch en route to Ecbatana: ‘Then Tobiah went down to bathe his feet in the Tigris river. But he let out a yell when a large fish suddenly leaped out of the water and tried to swallow the boy’s “foot” ‘ (6:3). Yet Raphael does not intervene to help Tobias land the fish, he merely issues instructions. Divine providence may have provided the fish, but it is Tobit’s responsibility to employ the gift, and it is this kind of synergy of divine providence and human obedience that seems to be in view throughout the book rather than any direct divine interventions. If the theological context provided by the book as a whole lifts the healing of Tobit and the exorcism of Asmodeus above the merely medicinal or magical, it raises them to the level of the providential rather than the truly miraculous.

But before reaching any final conclusions on the healing and the exorcism, we should take a closer look at each of them in turn. Once the fish gall has been procured and brought back, the curing of Tobit’s blindness turns out to be almost mundane:

Tobit 11:10–13a Rs

Tobit 11:10–13 Rv

Tobit got to his feet and came stumbling out through the courtyard gate. Tobiah went up to him with the fish gall in his hand. He blew on Tobit’s eyes and, steadying him, said, ‘It will be all right, father’. Then Tobit applied the medicine, and it made them smart. Next, using both hands, he peeled away the white patches from the corners of his eyes.

Tobit started toward the door, and stumbled. But his son ran to him and took hold of his father, and he sprinkled the gall upon his father’s eyes, saying, ‘Be of good cheer, father’. And when his eyes began to smart he rubbed them and the white films scaled off from the corners of his eyes.

Admittedly the account is more straightforward in Rv than in Rs. In Rv it appears that Tobit completes his own cure by rubbing his own eyes once they start to smart. Rs probably means that it is Tobias who completes the cure by peeling the white patches from his father’s eyes (since, in the absence of any clear indication to the contrary, one must assume that the subject of the succession of verbs remains the same). The most important difference, however, is the omission by Rv of the detail in Rs that Tobias first blew on his father’s eyes. Concerning this feature of the account, Moore remarks: ‘It is unclear whether this blowing in GII [=Rs] was intended to drive out the evil spirit or to “activate” the ointment (as does, for instance, one’s blowing on a cut covered with iodine).’ The latter possibility would appear to be excluded by the fact that the blowing takes place before the application of the ointment (τὸ φάρμακον), but there is nothing in the previous narrative to indicate that any demons were involved in Tobit’s blindness, which is accounted for in explicitly naturalistic terms. On the other hand, the blowing can hardly be a mere accidental relic of some underlying folk-tale, since it is also explicitly mentioned at 6:9 (Rs), when Raphael first instructs Tobias in the uses of the organs of the fish. Of course, one can attempt other naturalistic explanations for this action than that proposed by Moore, for example that the breath was meant to soften up the patches prior to removal or to blow away any dust or other foreign matter in Tobit’s eyes; or one might simply accept Zimmermann’s observation that ‘Anciently, breathing or blowing on a patient was an attempted cure.’

Whatever the exact function of the blowing, there are clear differences between this story and the healing of blind men in the Gospels. Unlike the blind man of Bethsaida, for example, Tobit is not cured on the spot but must wait for his son to go on a long journey and return. Again, whereas in Mark it is Jesus who effects the healing, in Tobit the credit for the healing is diffused among a number of actors: God acts through Raphael who acts through Tobias, and it is not clear that anyone actually works a miracle. Moreover, the story in Tobit has a different function from those in the Gospels.27 Whereas in the Gospels the healing stories illustrate Jesus’ power and, not least in the case of healing blind men, often also serve some symbolic meaning in their wider context, in Tobit the healing story is told not so much for the sake of demonstrating anyone’s numinous power but rather in the service of a Deuteronomistic theory of rewards and punishments (a theory explicitly contradicted at Jn 9:2–3 in connexion with healing a blind man).

The exorcism of Asmodeus is a rather more complex issue. In the Gospel accounts of Jesus’ exorcisms, Jesus’ word of command is sufficient to expel the demon. In Tobit, however, a number of factors are involved: the fumigation of the demon by Tobias, the binding of the demon by Raphael, and the prayer of Tobias and Sarah for mercy and protection (8:1–4). Again, unlike the Gospel accounts in which the victims of demonic attack appear to be possessed in the narrow sense, it is not immediately clear precisely in what manner Asmodeus is afflicting Sarah. Thus R.H. Pfeiffer remarks:

The author of Tobit, intent on writing a deeply religious and highly edifying book, took considerable liberties with the folk tale … As a result, it is difficult to determine, for instance, which variant of the Dangerous Bride motif was utilized in the story of Sara: was she the wife of Asmodeus (Bride of the Monster motif), his captive (Ransomed Maiden motif), or the object of his courtship (‘the demon loves her’ [6.14, Gr. 6:15]) in the standard Greek text [Codd. B.A.] and the Old Latin)? Or was she a victim of demonic possession …? Or, again, was Sara a Poison Maiden …? The author obviously obscured this crucial point of the tale for his own purposes.

Immediately on citing this passage, William Soil goes on to remark, ‘But, as Pfeiffer’s own catalogue of possibilities would indicate, the precise nature of Sarah’s predicament is not a crucial point in the tale, but a highly variable one.’ The point is well taken: Sarah’s predicament is a narrative device that allows the author to portray God’s providence and Tobias’s courage, and not necessarily a condition that the author is at all interested in for its own sake. Nevertheless, there are features of the narrative that would seem to limit the possibilities more than Pfeiffer supposes, and these suggest that Asmodeus is probably best conceived as an invisible, jealous lover murdering his rivals rather than an invading spirit entering Sarah’s mind or body and actually possessing her.

It is true that, according to Rv, Sarah is reproached with strangling her husbands (Οὑ συνίεις ἀποπνίγουσά σον τοὺς ἄνδρας, 3.8), and in Rs a single maid emphatically tells Sarah that it is she who has killed her husbands (Σὺ εἶ ἡ ἀποκτέννουσα τοὺς ἄνδρας σον). Yet earlier in the same verse it is stated that ‘the evil demon Asmodeus had slain each of them’ (rsv; Ασμόδαυς τὸ πονηρὸν δαιμόνιον ἀπέκτεινεν αὑτοὺς, Rv; Ασμοδαῖος τὸ δαιμόνιον τὸ πονηρὸν ἀπέκτεννεν αὑτοὺς, Rs). If both statements were true, then one would have to suppose that Asmodeus took possession of Sarah in order to strangle her husbands (or, in Rs, kill them some other way perhaps) with her hands. That a young woman would be able to overpower and kill a man when in a state of possession would not itself be surprising given accounts of the supernormal strength of possessed persons elsewhere (e.g. Mk 5:4; Acts 19:16). Yet it is not actually stated that Asmodeus murdered Sarah’s husbands through her agency, and the maid’s accusation may well be due to her ignorance of the demon’s role. Indeed, neither Sarah nor any other member of her household displays any apparent knowledge of the demon’s existence. This is strange, given that at 6:14 Tobias blames the jealous demon, not Sarah, for the death of her husbands. Does Sarah’s apparent ignorance of Asmodeus’s existence make it more or less likely that he acts by possessing her? It seems strange that she could be wholly unaware of being possessed, and even stranger still that those around her should remain unaware of it, unless Asmodeus was a very subtle kind of demon indeed. On the face of it one might most naturally suppose that Asmodeus acts, not as the possessor of Sarah’s mind or body, but as an invisible free-standing murderer, so that all anyone actually sees is that Sarah’s husbands mysteriously die on their wedding night.

Sarah certainly displays no sign of having just been freed from demonic possession when Tobias drives Asmodeus away, neither does she behave like a demoniac at any point in the story. On the contrary, at 6:12 Raphael describes her as ‘sensible and brave and very beautiful’. This is hardly the description of a demoniac, and since Raphael is an angel sent from God (3:17; 12:20) his opinions are presumably reliable. Moreover at 3:10–15 Sarah engages in devout prayer, again not an activity one would normally associate with a demoniac, though of course she could be conceived of as suffering temporary bouts of possession at other times. Finally, in Tobias’s view Asmodeus does not harm Sarah directly (ὅτι αὑτὴν οὑκ ἀδικεῖ, 6:15); this is surely another indication that the narrative does not view Sarah as herself possessed.

It is true that at 3:9 Sarah is charged with beating her maids, and at first sight such an action (which she does not deny) seems so out of character with the portrayal of her elsewhere that one might suppose she would only beat her maids in the frenzy of possession. Yet her fellow sufferer Tobit, devout though he is, also displays his faults. In particular he has just wrongly accused his wife of stealing from her employer (or perhaps, worse, of acquiring immoral earnings, 2:13–14). It is precisely this accusation that leads to Anna’s reproach, which in turn so distresses Tobit that he prays for death. Sarah’s beating of her maids thus provides a similar occasion for the maid’s reproach, thereby giving a parallel structure to the accounts of the two sufferers and so linking their stories.

Some indication that Asmodeus is to be regarded as a jealous (demonic) lover is provided by Tobias’s statement that ‘a demon is in love with her’ (rsv). Admittedly, this statement is lacking in Rs, but it is present not only in Rv but also the Old Latin, and it appears to be supported by 4Q196. Further support for the idea that Asmodeus is behaving like a jealous husband who needs to be divorced from his unwilling spouse is supplied by the description of what Asmodeus has been sent to do at 3:17: καὶ ἀπεστάλη Ραφαηλ ἰάσασθαι τοὺς δύο … καὶ λῦσαι Ασμοδαιον τὸ δαιμόνιον τὸ πονηρὸν ἀπʼ αὑτῆς. The text here does not say that Asmodeus is to release Sarah from Asmodeus, but that he is to release Asmodeus from Sarah. As Paul-Eugene Dion remarks, ‘One hardly expects that this “evil demon” needs to be “loosed” from Sarah; it is rather this poor girl who needs to be loosed from such a partner!’ Dion goes on to suggest that the verb λῦσαι was equivalent to the Babylonian Aramaic paṭṭar, part of the legal language of divorce amongst Jews. But Dion also points out that ‘we know of not less than 13 incantations inscribed on bowls where the exorcist claims to pronounce the divorce between a demon and his client’. Admittedly these are of much later date (seventh century ce), but Dion argues that there is a long tradition of Babylonian magic going back many centuries before Tobit and that the exorcism of Asmodeus reflects this tradition. The redactor of Rv, he suggests, was unfamiliar with this type of language and so changed the obscure (for him) λῦσαι to δῆσαι, which describes what Raphael goes on to do in 8:3.

One might suppose that the binding of Asmodeus reflected Jewish ideas about the intermediate state of defeated demons prior to their eschatological punishment, as at 1 En. 10.4, where it is again Raphael who is said to do the binding. Here again, though, Dion shows that the motif of binding demons would be quite at home in the tradition of Babylonian magic.38 To be sure, these explanations are not mutually exclusive, since Jewish thought could easily have borrowed the notion of binding demons from the surrounding culture and then developed it in its own way. One effect of Dion’s argument, however, should be to caution against too readily assuming that the details of the exorcism in Tobit represent a particularly Jewish way of doing things. Moreover, the use of the organs of the fish to fumigate the demon allows the fish to serve in the rescues of both Sarah and Tobit, thus once again providing a close link between their two stories in a manner that nicely illustrates the workings of divine providence. It may well be this narrative convenience, rather than any particular desire to promote Babylonian exorcism techniques, that made the author of Tobit relate the exorcism in these terms,

Even if Tobit has borrowed a folk tale containing a thoroughly Babylonian-magical approach to exorcism, features of the text indicate that the procedure has been effectively Judaized. The stench of the burning fish organs is enough to drive Asmodeus all the way to the desert of southern Egypt, but presumably the point of both the prayer and the binding is that it would not be enough to keep him there (cf. Mt. 12:43–45). Tobias cannot know that Raphael has chased after Asmodeus to bind him, since at this point in the narrative the angel is known to Tobias only as his human kinsman Azariah. The point of the prayer that follows (at least in Rs, which explicitly mentions σωτηρία as one of the things to be prayed for) is then to secure divine protection against the demon’s possible return. In fact, this part of the prayer has already been answered by Raphael’s action in binding Asmodeus hand and foot (8:3). Although this appears to be out of sequence (the prayer is uttered after its being answered is described), it would not be out of keeping with the dramatic irony the author of Tobit goes in for (cf. Raguel’s digging a grave for Tobias at 8:9, when the reader already knows for certain that it will not be needed). The magical procedure is thus theologized, and so turned into a miracle, by having Tobias act as PNP in requesting divine aid and Raphael as MNP in providing that aid.

As we have already noted, the whole procedure is quite different from the exorcisms performed by Jesus according to the Gospels. There, a word of command is apparently sufficient to banish the demon for good, and (so far as one can tell) the form of demonic attack that Jesus is countering is possession in the narrow sense, which does not seem to be in view in Tobit. Again, in Tobit, no eschatological significance is attached to the exorcism (unless this is hinted at by the parallels to the Enochic literature in which the lust and binding of fallen angels both feature), and there is a sense in which, unlike the exorcism stories in the Gospels, that in Tobit is something of an anticlimax. Nevertheless, by incorporating this account in his story and giving it a clear theological interpretation, the unknown author of Tobit suggests that an exorcism was something that a godly Jew of the second or third century bce could honourably and usefully perform.

It also shows that delivery from blindness and demonic attack is something God could be conceived as wanting to bring about—at least for pious people who trust in him. Furthermore, Tobit and Sarah have in common with people Jesus heals in the Gospels that they are ‘little people’, not powerful rulers or great national heroes. But it should be noted that in the Book of Tobit, no one character performs quite the role that Jesus performs in the healing stories of the Gospels. The role of healer/exorcist is shared between Tobias and Raphael. Yet several factors (the angel’s name, his superior expertise, his having been sent from God specifically to heal, his completing the defeat of the demon, and the fact that there is no indication that Tobias makes a career of healing) suggest that it is Raphael more than Tobias who resembles Jesus as healer. It is striking in this connexion how closely the language Raphael uses of himself once he has revealed his true identity resembles some of that Jesus uses of himself in the Fourth Gospel, most notably, ἰδοὺ ἐγὼ ἀναβαίνω πρὸς τὸν ἀποστείλαντά με (Tob. 12:20).


The Jewish apologist Artapanus wrote his work On the Jews some time in the third, second, or first century bce. The three surviving fragments are preserved in Eusebius’ s Praeparatio Evangelica, Eusebius having extracted them from the work of Alexander Polyhistor. Part of the third fragment (on Moses) is also paralleled in the Stromata of Clement of Alexandria. The original language of Artapanus was almost certainly Greek, as is the language of the surviving fragments.

The first fragment (Praep. Evang. 9.18.1 = Art. 18.1) gives a brief explanation of the name of the Jews/Hebrews and a notice of Abraham’s stay in Egypt. The second (Praep. Evang. 9.23.1–4=Art. 23.1–4) is a brief account of Joseph in Egypt. The third and most substantial fragment (Praep. Evang. 9.27.1–37=Art. 27.1–37) contains a romanticized account of Moses’ career. Neither of the first two fragments has anything miraculous, but the third narrates a number of biblical and non-biblical miracles. It is with this third, Moses, fragment that the discussion will accordingly be concerned.

Artapanus‘s account of Moses falls into two sections. The first has very little direct contact with the biblical narrative, and is mainly concerned to present Moses as a great benefactor to Egypt, the originator of most of Egyptian culture (including Egyptian religion), despite being unjustly persecuted by the envious Egyptian ruler Chenephres. Chenephres’s plot to have Moses killed leads to Moses’ instead killing his would-be assassin and hence to his flight from Egypt. From then on, in the second section, Artapanus gives a compressed account of the exodus narrative, with some additional miracles. The purpose of this composition was probably to counter Egyptian anti-Jewish propaganda in writers such as Manetho by presenting the Jews as a blessing rather than a bane to Egypt, to represent the Jewish national hero Moses as outperforming the gods (e.g. Isis, Osiris and Thot-Hermes) and heroes (e.g. Sesostris) of the Hellenistic-Egyptian world, and also perhaps to encourage a Jewish community in competition with local cults. The concentration on matters Egyptian not only in the Moses fragment but in the other two as well strongly suggests an Egyptian provenance for the work.47

Artapanus‘s miracles all occur in the second section of the Moses fragment. They begin with Artapanus‘s version of the burning bush, a version that in fact dispenses with the bush in order to heighten the miraculous by insisting that ‘fire was suddenly kindled from the earth and it burned although there was no wood or other kindling material in the place’ (Art. 27.21). When Moses returns to Egypt to confront the king,49 he is thrown into prison, but during the night the prison doors open αὑτομάτως and Moses escapes. Moses enters the royal chamber, and the king mockingly challenges him to pronounce the name of his god. Moses whispers the divine name in the king’s ear, and the king collapses speechless until Moses revives him. Moses also writes the divine name on a tablet; a priest disparages the contents of the tablet and duly dies in convulsions.

Thereafter Artapanus broadly follows the biblical story with an account of the plagues. In Artapanus, however, these are worked entirely by Moses and his rod; unlike the biblical account, and unlike some of the other retellings of the biblical account examined above, neither Aaron nor Yahweh is given any role to play. Moreover, the plagues briefly described by Artapanus are not entirely identical with their biblical counterparts; for example the plague of hail is accompanied by earthquakes, and the death of the Egyptian first-born is omitted, along with the cattle-disease and the darkness over the land. Following the earthquakes and hail the king lets the people go, but then pursues them to the Red Sea. Artapanus provides two explanations for the Red Sea crossing, including a rationalistic one, attributed to the Memphites, involving tides, but again heightens the miraculous by having the Egyptians destroyed by fire as well as water. Finally, there is a brief notice that God rained down food for the Israelites in the desert, and Artapanus’ account (or the surviving extract of it) is brought to a conclusion with a description of Moses’ appearance.

From the point of view of the current investigation, perhaps the most striking feature of this account is the way in which Moses is apparently transformed into an immanent BNP (or perhaps, simply into a magician). As Tiede correctly observes,

Nor is Artapanus particularly cautious about maintaining that Moses is subservient to God in all that he does. After a very brief mention of his initial fright before the miraculous fire (27.21), Moses ‘took courage and determined to lead a hostile force against the Egyptians’ (21.21) [sic—the actual reference should be 27.22]. To be sure, he is acting on command of the ‘divine voice’ now further identified with ‘the Lord of the world’ (27.22), but Moses runs his own show until the ‘divine voice’ commands him to divide the Red Sea (27.36).

Such a presentation of Moses has little direct bearing on the miracles of the historical Jesus, but may provide a significant parallel to the way in which Jesus’ miraculous activity is presented in the Gospels. If Artapanus can turn Moses into an immanent BNP for the purposes of religious propaganda in a Hellenistic environment, this might provide a parallel to what the Evangelists (or their tradition) did with the miracle stories about Jesus.51 The question then is whether Artapanus does in fact represent Moses as an immanent BNP (as opposed, say, to a magician or MNP), which in turn raises the question how far Artapanus has a clear concept of miracle at all.

One may begin by examining who precisely is said to be responsible for the various miracles of the Moses fragment. In one or two places it is clearly God. When ‘Moses prayed to God that he might thereupon give the people an end to their sufferings’ a fire was ‘suddenly kindled (ἀναφθῆναι)’ and a ‘divine voice’ instructed Moses to liberate the Jews from Egypt (27.21). Here ἀναφθῆναι must be taken as a divine passive, with Yahweh the clearly implied agent. With one possible exception (the death of Chenephres) this is the first miracle narrated. The final miracle narrated also has God as the explicit BNP, namely when ‘God rained for them meal like millet’ in the wilderness (27.37). It may be significant that Moses’ miraculous activity is thereby framed by two activities that place his perhaps otherwise magical acts in a theistic context. This bears some similarity to the way in which the Gospel accounts frame Jesus’ mighty works between an introduction (the baptism or prologue, and in two cases, birth narratives) and a conclusion (the resurrection) that likewise show that Jesus is acting as an agent of God.52 But Artapanus lacks any real equivalent to the features in the Gospels that keep the central character closely bound to God, such as the synoptic transfiguration scene or the repeated Johannine references to the relation of the Son to the Father.

In several other places in Artapanus the BNP may well be God, but this is not explicitly stated. For example, when Moses is imprisoned, the prison doors are said to open αὑτομάτως (27.23). At Ant. 4.47–48 Josephus regards things happening αὑτομάτως as an alternative (and incorrect) explanation to their happening by the divine will, but there is no reason to suppose that Artapanus is following the same usage. His point is that the doors open by no human or other visible agency. In the absence of any evidence to the contrary, then, one is presumably meant to understand that it is God who opens the doors (and incapacitates the guards) for Moses. This is certainly how Clement of Alexandria seems to have taken it. In what follows the king mockingly asks Moses for the name of the God who sent him, and is struck dumb when Moses utters the divine name in his ear. The same divine name proves fatal to an Egyptian priest who disparages it after Moses has inscribed it on a tablet (27.24–26). This could be taken in more than one way. One might argue that since it is the name of God, who sent Moses, that causes these effects, it is God’s power that is at work all along. On the other hand, one might equally well argue that since the name of God suffices to stun Pharaoh and kill the priest, Moses is employing it as a magical word of power. A decision between these alternatives must await discussion of the other material in Artapanus.

There are two other places where the reader is left to suppose that the BNP is probably God. The first is the notice that ‘Chenephres became the first of all men to contract elephantiasis and he died’ (27.20).55
Artapanus goes on to explain that ‘He encountered this fate because he had enjoined the Jews to wear linen garments’ so that they were readily identifiable (27.20). One therefore assumes that his novel death was a divine punishment. The second is rather different. In J. J. Collins’s translation, at the Red Sea a ‘divine voice’ commands Moses ‘to strike the sea with his rod and divide it’ (27.36). Given what Moses has already achieved with his rod (more on which below), this makes it hard to decide whether Moses is acting as MNP with God as BNP, or whether God is commanding Moses to act as BNP (or magician). But Collins’s translation results from a conjectural emendation from διαστῆναι (‘stand apart’ or ‘go away’) to διαστῆσαι (‘divide’). This emendation hardly seems necessary, since it would make reasonably good sense for Moses to be commanded to ‘stand aside’ after smiting the sea if something dramatic was about to happen there. A further complication is that not all manuscripts of Eusebius’s Praeparatio Evangelica even contain the words καὶ διαστῆσαι at this point. Given also the possibility that either Eusebius or Polyhistor may be paraphrasing, abbreviating or simply misquoting Artapanus, it becomes doubtful whether one can base very much on the exact wording here. It is reasonably clear that Moses strikes the sea with his rod at the divine command, whereupon the sea divides. Whether it is God or Moses’ rod that actually effects the division of the sea is less clear.

As has already been observed, up to this point Moses has performed a number of miracles with his rod without explicit divine aid. The sequence begins with the king asking Moses to perform a sign (σημεῖόν τι) for him. Moses obliges by casting down his rod and making it a serpent. When this terrifies the onlookers, he picks it up and it turns back into a rod (27.27). Moses then strikes the Nile with his rod and it floods the land, thereby originating the annual inundation. In a rationalizing touch, the water does not turn to blood, but becomes stagnant so that the people cannot drink it (27.28). The king promises to release the people in a month’s time if Moses restores the river, which Moses duly does by once again striking it with his rod (27.29). At this point the king summons some priests and commands them to perform some marvels (τερατουργήσωσί) under threat of death and the destruction of their temples. ‘They then, through some superstitious tricks and charms (διά τινων μαγγάνων καὶ ἐπαοιδῶν), made a serpent and changed the color of the river‘ (27.30). This is sufficient to persuade the king to renew his hostility to the Jews, whereupon Moses brings about a sequence of plagues. First he strikes the ground with his rod to release winged creatures to afflict the Egyptians (27.31). Then he releases a frog through his rod together with lice and locusts (27.32). Finally, Moses brings about hail and earthquakes, which kill many people and destroy all the houses and most of the temples (27.33). At this point the king releases the Jews (27.34). The plagues thus occur in three groups: the insects that afflict the Egyptians with sores are summoned by Moses striking the ground with his rod; the frog, lice, locusts and lice are merely released by Moses through his rod (although striking the ground is implied later in the verse—see below); and the hail and earthquakes are brought about by Moses with no explicit reference to the rod at all.

Although Artapanus describes the wonders of the Egyptian priests as being wrought ‘through some superstitious tricks and charms’, he does not sharply distinguish their merely magical technique from Moses’ divinely empowered miracles as does, for example, Josephus. On the basis of the evidence examined so far, one might suppose that Moses is being portrayed simply as a superior magician, albeit a magician operating broadly under divine authority. One might alternatively regard him as a divinely commissioned BNP, since no source of numinous power other than Moses and his rod features in this sequence of miracles. Or one might begin to suspect that for Artapanus the distinction between magician and immanent BNP is one without any real difference, since for this author what matters is the impressiveness of the wonder rather than any well-defined theology of miracle.

There remains one significant piece of evidence yet to be considered, however. The verse describing the plague of frog, locusts and lice also assigns a role to Isis:

Again Moses released a frog [sic], through his staff, and in addition to these things, locusts and lice. On this account the Egyptians dedicate the rod in every temple and similarly (they dedicate it) to Isis, since the earth is Isis, and when it was struck it released the marvels (τὰ τέρατα) (Art. 27.32).

In this passage the (reluctant) BNP behind the plagues appears to be Isis, compelled to perform marvels when struck by Moses’ rod.60 A strict application of my definition of magic would thus make Moses a magician here, since Isis is not the God regarded as legitimate by Artapanus. But in this case such a strict application would not be appropriate. For Artapanus the point is not whether Moses’ source of power was legitimate, but that Moses overmasters Egyptian religion. The inferiority of Egyptian religion is also brought out at 27.33 with the collapse of ‘most of the temples’, which is itself an ironic fulfilment of the king’s threat at 27.30 to tear down the temples of the priests if they failed to match Moses’ marvels. The point is that Moses thereby demonstrates his superiority, and thus that of the cult he founded, over any Egyptian cult.63

To attempt to force a fully coherent picture out of the surviving fragments of Artapanus would probably be a vain effort. On the one hand Artapanus has Moses introduce several innovations into Egyptian culture and religion, presumably to counter charges that the presence of the Jews had proved harmful to Egypt. On the other, Artapanus has Moses demonstrate his complete superiority over Egyptian religion, so that Moses ends up opposing what he helped to found. This tension presumably comes about because Artapanus has now turned to fight on a different front. Artapanus’s presentation of miracle seems similarly ad hoc. On the one hand Moses’ miraculous activity starts when God enters the fray, and is set within a framework of divine miraculous action. Moreover, Moses’ actions against the Egyptians on behalf of his people are set within the general framework of obedience to the commands of a divine voice. On the other hand, Artapanus is not particularly scrupulous in ensuring that God remains the BNP for the miracles that follow, and instead creates the impression that Moses is quasi-magician, quasi-autonomous BNP. Perhaps the notice at 27.36 that Moses struck the sea with his rod in obedience to the divine voice is meant to bring his previous rod-wielding under divine auspices, but by then the rod has also been associated with Isis. The general impression that Artapanus wishes to create with all this seems clear enough: Moses, the servant of God, proves mightier than Egyptian religion, and so the Jewish cult is superior to the Egyptian cults. But to try to extract a clearly defined theology of miracle from all this would be a hopeless task.

That a Jewish author can thus turn Moses into something resembling an immanent BNP remains a potentially interesting parallel to the Gospel presentation of Jesus, provided the parallel is not pushed too far and the differences are also respected. The Jesus of the Gospels is not presented as in competition with or ambivalent relationship to pagan cults. The dynamic of Jesus’ relationship to the Jewish authorities on the one hand and demonic powers on the other is thus somewhat different from that of Moses’ relationship to Egyptian authorities and cults. In particular, Jesus is represented neither as the original benefactor nor as the triumphant conqueror of the major religious tradition with which he comes in conflict, but as its true fulfilment (albeit a fulfilment that may prove as destructive to temple buildings as Artapanus’s Moses). Neither do the Gospels flirt with polytheism in the manner that Artapanus appears to risk; in the Gospels the only supernatural powers operating apart from God (and his angels) are purely demonic.

A closer parallel between Artapanus and the Gospels may lie in their competitive historiography. ‘In an age when ethnic groups bolstered their pride with tales of national heroes, Artapanus portrayed Moses as the greatest hero of them all.’ Just as Artapanus’s, Moses has to outperform Sesostris, so the Evangelists’ Jesus outshines Moses and Elijah. But again the relationship between Jesus and the great Jewish heroes of the past is not so straightforwardly one of competitive replacement as seems to be the case with Artapanus’s Moses and the Hellenistic-Egyptian figures he is set to outdo. In both Artapanus and the Gospels one may detect a tension between positive and negative attitudes towards a competing religious tradition, but in the Gospels the resolution lies in the direction of appropriation, whereas in Artapanus it veers towards repudiation.

It cannot thus be assumed that the tensions that resulted in Artapanus’s portrait of Moses would necessarily be operating in the case of the Gospels. Artapanus shows that a Jewish author writing in Greek could be capable of turning a Jewish hero into something like an immanent BNP, but that is not the same thing as showing that the presentation of Jesus as a BNP is solely or necessarily the result of presenting the Gospel in a Hellenistic environment.


This chapter has examined two rather different prose narratives that possess some potential affinities with the Gospel accounts of Jesus’ miracles. It turns out that the affinities are not all that close. The healing story in Tobit is not all that miraculous, and the exorcism is scarcely the reversal of the type of possession envisaged in the Gospel stories, neither is it carried out by similar means. The miracles performed by Moses in Artapanus’s account are not particularly similar to those of the Gospels’ Jesus, and its euhemeristic account of Moses’ benefactions to Egyptian society create a tension with its subsequent portrayal of Moses quite lacking in the narratives about Jesus. Yet some affinities remain. The account of the exorcism in Tobit at least shows that exorcism is an activity that could be attributed to a good and devout Jew, and its echoes of the Enochic tradition provide a further hint of how an individual exorcism story might be placed in a broader eschatological perspective. Again, for all the differences between Artapanus and the Gospels, it cannot be denied that they share the unusual trait of presenting their central characters as immanent BNPs (at least to the extent that Artapanus’s Moses can be characterized thus and not simply as a magician). These two texts may thus provide part of the Jewish context for understanding the Gospel accounts of Jesus’ miraculous activity, even if one doubts their direct bearing on the ministry of the historical Jesus.

We have now examined a wide range of Second Temple texts for Jewish views on miracle, and in doing so we have probably covered the bulk of the most significant material. There are, of course, many other texts in which miracles occur, or are alluded to, or which might shed some further light on the demonological or eschatological context of miracle. But the returns from treating all these other texts in detail would rapidly diminish, since in none of them does miracle play any major role. Moreover, to continue piling detail upon detail would be to risk confusing the picture rather than clarifying it. At this point it is best to step back to look at the picture as a whole, perhaps allowing the other texts to supply the odd additional detail here and there. The next chapter, therefore, will summarize the findings from the material already discussed while briefly introducing some further evidence from other relevant material.

Chapter 9

Miracle in Second Temple Literature

General Observations

The previous seven chapters have examined a variety of Second Temple literature for its views on miracle. This chapter will attempt to draw together the threads of the previous seven by conducting a thematic overview, surveying first the agents and then the purposes of the miraculous in this literature as a whole. This survey will also introduce evidence from some other texts among the Apocrypha and Pseudepigrapha, although there will not be the space to deal with them in any detail.

The texts surveyed here come from a variety of dates and social settings, and appear in a variety of genres with a variety of aims. Moreover, some of the additional texts to be considered have clearly been subject to Christian interpolations. To lump all these writings together risks making a false synthesis, as if they all ‘without regard to context and circumstance, testify to one and the same Judaism’. Indeed, the last seven chapters have already exhibited a wide variety. The aim here is thus not to produce a synthetic account of Jewish ideas of miracle around the time of Christ. It is rather to see what ideas about miracle were or were not current in this body of literature. But first one may make some general observations on the role of miracle in this literature as a whole.

First, this literature displays little preoccupation with the miraculous. The apparent exceptions must be treated with caution; for example The Lives of the Prophets is particularly interested in the miracles the prophets performed, and The Testament of Solomon relates bizarre feats the king accomplished through his control of demons, but neither of these texts can safely be taken a straightforward reflection of Second Temple Judaism. Again, although the literature exhibits a lively belief in demonic powers, exorcisms are mentioned only rarely. Most of the Jews who produced Second Temple literature were more interested in keeping the law, or in the history or prospects of the nation, or in wisdom and right living, or in the revelation of heavenly realities, or in promoting their own party or people and condemning their enemies. Miracle might play a role in all of these things, but for the most part it was only a minor one. Moreover, it is not even clear that for most of these authors (Philo and Josephus excepted) ‘miracle’ formed a clearly distinct category of divine action.

Secondly, where this literature does show an interest in miracle it is usually in the biblical miracles of the past, sometimes with legendary elaborations. Not surprisingly, the miraculous events surrounding the exodus receive particular attention, as in Philo (especially De Vita Mosis), Josephus (Ant. 2–4), Wis. 10–19, Artapanus, Ezekiel the Tragedian, Jannes and Jambres,
Jubilees 48 and scattered references elsewhere. In some cases further miracles are added to the biblical account, for example at LAB 27.9–11; 35.6–7, and also at Art. 27.23–26 and 27.36–37. One also finds a tendency to attach additional miracles to famous biblical figures, particularly in the Testament of Solomon, the Lives of the Prophets, and 4 Baruch.

Thirdly, outside Josephus, stories of post-biblical miracles are rare. 2 Maccabees gives a notably miraculous account of the Maccabaean war in contrast to 1 Maccabees, and both 3 Maccabees and 4 Maccabees narrate miraculous interventions on behalf of the sanctity of the Jerusalem temple, as well as 3 Maccabees‘ account of the Alexandrian Jews’ deliverance from Ptolemy’s elephants, but none of these works narrates events that are directly contemporary with their authors.

Finally, there are only two types of passage that come close to describing anything like the miraculous career of Jesus. The first is plainly of Christian provenance, since it contains summaries of Jesus’ miracles thinly disguised as prophecies, for example Sib. Or. 1.351–59; 6.13–16; 8.272–78; T. Adam 3.1; Odes 39:9–11. The second is formed by passages describing the coming of Beliar (or some equivalent anti-God figure). Again, two of the relevant passages come from the Sibylline Oracles (2.167–69; 3.63–70). These are reminiscent of the warnings against false prophets and false messiahs in Mk 13:22. The most striking text of this kind is plainly Christian in its present recension; Apoc. Elij. 3.5–13 lists precisely the kind of miracles attributed to Jesus, but attributes them to the ‘son of lawlessness’, concluding:

He will multiply his signs and wonders in the presence of everyone. He will do the works which the Christ did, except for raising the dead alone. In this you will know that he is the son of lawlessness, because he is unable to give life.

Particularly striking is the fact that the very signs that Jesus uses to answer the Baptist’s question at Mt. 11:5 are here used to identify the ‘son of lawlessness’. The final paragraph, which says that the son of lawlessness will be unable to raise the dead as Christ did (contrast Sib. Or. 3.66), looks suspiciously like a Christian attempt to neutralize the sting of this passage, which may suggest that part of it was pre-Christian. As we have seen, 4Q521’s apparent prediction that the Messiah will perform miracles like those described in Mt. 11:4–5 should probably not be read this way.

It seems likely that the texts discussed in the last few paragraphs bear witness to beliefs about a Beliar or ‘Antichrist’ figure that were current in the first century ce independently of Jesus. If that is so, then the ironical conclusion from this preliminary survey must be that the figure in the Pseudepigrapha and Apocrypha that Jesus’ miracle-working makes him most resemble is that of Beliar.

Agents of the Miraculous

a. God

That God is the real BNP may often be assumed even when it is not explicitly stated.5 Here, however, we are concerned with miracles that are attributed to God without explicit mention of any intermediary.

We have already seen that for Josephus and Philo, a miracle is essentially an act of God, even when a human intermediary is involved. In Philo, miracles seldom do involve any human intermediary, Moses and Aaron being the only real exceptions. Pseudo-Philo and Jubilees also tend to stress God’s action and only seldom give any human actor any role in miracle-working. The agent of the miraculous in the Wisdom of Solomon is likewise nearly always God (often through wisdom). Ben Sira’s ‘Praise of Famous Men’ tends to praise men more than God, but nevertheless credits God directly with the miraculous destruction of Dathan, Abiram and Korah for the sake of Aaron (45.19), while the reference to the destruction of Sennacherib’s army shows that there may not have been much difference in practice between attributing something to God and attributing it to an angel sent by God: ‘The Lord smote the camp of the Assyrians, and his angel wiped them out’ (48.21). Elsewhere in Ben Sira miracles tend to be attributed directly to God.

Again, as we have seen, the Qumran literature makes occasional brief allusions to the great saving acts of God. Although the fragmentary nature of both the Prayer of Nabonidus (4Q PrNab) and the Messianic Apocalypse (4Q521) makes them hard to interpret, it may be that both these texts also attribute any miracles to which they refer directly to God; the precise role of the anonymous Jew in the former text and of the Messiah in the latter is, however, hard to fathom. It is also hard to be precise about the various demonological or apotropaic texts found at Qumran, again in part due to their fragmentary nature. Some of these (especially 4Q560) appear to be magical in nature; others (especially 4Q510 and 511) could be seen as threatening demons with God’s power, and hence as presupposing that God is the real and direct worker of any miraculous counter-measures against the demonic foe.

For further evidence, one might turn first to apocalyptic texts. In 4 Ezra, the ‘wonders and mighty works’ that manifest the beginnings, and the ‘requital and signs’ the end, appear to be attributed directly to the Most High. It is also the Most High who is said to perform signs for the returning ten tribes, enabling them to cross the Euphrates (13:44) and to show wonders to the survivors of the judgment upon the nations (13:50). So far as one can tell, the miraculous happenings in 2 Baruch are also attributed directly to God. The miraculous abundance described in ch. 29 is associated with the appearance of the Anointed One, but it is not attributed to him. Where there is a general cosmic upheaval before the end, it is not always clear whether this is to be attributed to the judgment of God or the final fling of the demonic powers. Moreover it is sometimes unclear whether this language should be regarded as ‘miraculous’ at all. The description of the appearance of God’s kingdom in T. Mos. 10 makes it clearly the work of God; in this passage the cosmic upheaval functions as part of a theophany—creation quakes before the presence of its Creator. His acts then include the destruction of the evil powers (10.1). It also appears to be God who is expected to heal his servants and destroy Satan at Jub. 23.29–30.

Several other Second Temple texts attribute miracles directly to God. At 3 Macc. 5:27–28 it is God who makes Ptolemy forget his order to have his elephants trample the Jews to death. The ‘Hellenistic Synagogal Prayers’ also tend to attribute mighty saving acts directly to God, e.g. Hell. Syn. Pr. 4.24; Hell. Syn. Pr. 5.11–12; Hell. Syn. Pr. 12.73–76; and Hell. Syn. Pr. 7.6. Baruch 2:11 recalls the exodus as the saving act of God. Artapanus wavers between attributing these miracles to God and attributing them to Moses. It is, however, God who is said to be solely responsible for the manna miracle (Art. 27.37), and it is presumably God who opens the prison doors for Moses to facilitate his escape (Art. 27.23). The book of Daniel, together with its apocryphal additions (The Prayer of Azariah, Bel and the Dragon) narrates miracles God works on behalf of the heroes, rather than miracles worked by them. 1 Maccabees is restrained towards the miraculous, but it appears to be God who strikes Alcimus down for daring to order the demolition of the sanctuary wall (1 Macc. 9:54–56). 2 Maccabees is full of angelic interventions, but at 9:5 God acts on his own initiative to punish Antiochus directly with a painful and fatal illness.

b. Quasi-Divine Agents

Only in the book of Wisdom is Sophia represented as an agent of the miraculous (on which see Chapter 4 above). Angels appear more frequently as agents of the miraculous. They are often employed as the agents of judgment or restraint on the devil and his minions (e.g. 1 En. 10.4, 11–16; 40.7; Jub. 10.7–9; T. Sol. 1.6). They are also common as miraculous warriors fighting on behalf of God’s people (LAB 27.10; 1 Macc. 7:41; 2 Macc. 10:29; 11:8; 15:22; Sirach 48:21) or defending them from attack (2 Macc. 3:25; 3 Macc. 6:18–19; 4 Macc. 4:10). They also effect the miraculous deliverance of the devout from deadly peril (LAB 38.3–4; Dan. 3:25; 6:22; Prayer of Azariah 26; Bel and the Dragon 33–36). At Jub. 48.10–13 an angel assists Moses against Mastema. Raphael, in accordance with his name, appears in more than one place as the agent of healing (1 En. 40.9; Tob. 3:17; 6:3–5; 8:3). Angels are associated with sundry miracles at T. Abr. 3.11, LAB 26.4 and LAB 35.7. There may not be a great deal of difference in attributing a miracle directly to God and attributing it to an angel acting under his orders; these angels may often be more literary devices for expressing the workings of the transcendent God within the world than independent agencies.

c. Human Agents

We have already seen that roughly 60 per cent of the miracles in Josephus’s Antiquities are associated with prophetic figures (including Moses) and that Ben Sira’s ‘Praise of Famous Men’ tends to associate prophets and miracles. It turns out that in other Second Temple literature prophets (including Moses) also form by far the largest class of human miracle-workers, and that they tend not to be portrayed as BNPs. But we shall also see that whereas Kahl’s terminology continues to be useful, in some cases it starts to break down.

A case in point is The Lives of the Prophets. On the whole, this text is careful to present the miracle working prophets as PNPs. This is so of both the miracles attributed to Isaiah (1.2, 3), of Jeremiah’s ridding the Egyptians of asps and crocodiles through prayer (2.3), and of many of the miracles in the summaries of the deeds of Elijah (21) and Elisha (22). Elijah is additionally portrayed as an MNP at 10.6 (raising the widow’s son) and 21.6 (keeping the widow fed). There are, however, several places where prophets are said to perform miracles without any explicit mention of God. At 2.11 Jeremiah seizes the ark of the Law and causes it to be swallowed up by a rock, apparently on his own initiative, but here the general context, together with the fact that Jeremiah is clearly portrayed as a PNP at 2.3, probably suggests that he should be not seen as a BNP. The same cannot be said of Habakkuk at 12.6–7; here, the prophet simply announces to his family that he is about to visit a far country, and then goes and delivers a meal to Daniel. Neither divine nor angelic activity is mentioned in the text.

In ch. 21 Elijah performs one miracle without apparent divine assistance, striking the Jordan so that it is divided and he and Elisha can cross dry-shod (21.14). But given this chapter’s tendency to mention Elijah’s prayer elsewhere, this can probably be regarded as a slip rather than an attempt to portray Elijah as a BNP. The treatment of Elisha in ch. 22, however, is a different matter. Here Elisha is said to perform a number of miracles without any explicit reference to God (22.4, 7, 8, 10, 12, 17). On the other hand Elisha is also said to invoke God or pray to perform a number of others (22.6, 9, 11, 16). Given the large number of miracles listed for this prophet, it would be reasonable to suppose that the apparent BNP miracles here are more a stylistic variation to avoid repeating ‘he prayed’ than an oscillation of theological opinion. In all probability, then, Elisha should be seen as a PNP rather than a BNP here, despite apparent lapses.

Ben Sira’s ‘Praise of Famous Men’ also recalls a number of miracles performed by the prophets, and, as we have seen, there is an apparent tendency to turn some of them (Moses, Joshua, Elisha and Isaiah) into immanent BNPs, though this should probably be regarded as poetic license rather than deliberate thaumatology. Ben Sira 48:10 (possibly an interpolation) refers to Elijah’s eschatological role. The only place in this literature where a prophet (again Elijah) is given any kind of miraculous eschatological role is at Sib. Or. 2.187–89 where it is said that he will drive a chariot from heaven and perform three signs at the end. But an Elijah descending from heaven in a chariot is scarcely an ordinary mortal. There is no particular reason to suppose this part of the oracle to be the work of a Christian redactor, and the idea may well be derived from Mal. 4:5.

Another miracle-working figure to whom considerable attention is given in this literature is Moses. Whether or not the Pentateuch intends to depict him as a prophet, Deuteronomy appears to consider him one (Deut. 18:15, 18; 34:10), and it is reasonable to suppose that many Jewish authors will have shared that view (e.g. Wis. 11:1; T. Mos. 3.11). We have already seen that some of the texts attribute the exodus events directly to God. We must now look at those that emphasize Moses’ role.

We have already seen that the miracles associated with Moses constitute roughly 40 per cent of those narrated in Josephus’s Antiquities, and although Philo does report some miracles worked through Aaron, Moses is the only human figure whose performance of miracles Philo makes anything of. Both these authors nevertheless subordinate Moses’ miracle-working activity to God; there can be little doubt for either of them that Moses is no more than PNP or MNP and that it is God as BNP who actually performs the miracle. The same is true of Pseudo-Philo and Jubilees, both of which narrate some of the miracles associated with Moses, and both of which appear to attribute a greater role to Moses in their general statements than in their more detailed account of events.

Artapanus turns the story of the burning bush into a miracle wrought by Moses as PNP (Art. 27.21). On being requested by Pharaoh to perform some signs, Moses subsequently throws down his rod so that it turns into a serpent and causes the Nile to flood at the stroke of his rod (27.27–28). It is also Moses who brings the plagues on Egypt (27.31–33). Since God is not mentioned in these passages, Artapanus comes close to representing Moses as a BNP here. But at the Red Sea Moses is reduced to an MNP once more, since he strikes the water at God’s command, although it is this action that is said to divide the sea (27.36). The manna miracle, on the other hand, is attributed directly to God (27.37). Overall, Artapanus comes closer than anyone else to turning Moses into an immanent BNP.

The surviving fragments the Exagōgē of Ezekiel the Tragedian tend to represent Moses as an MNP. Lines 120–31 contain God’s commands to Moses on the employment of his rod and hand to perform signs before Pharaoh. At lines 132–74 God similarly instructs Moses on the use of his rod to bring about the plagues. The sole Egyptian survivor of the Red Sea miracle attributes the parting of the waters to Moses’ rod once more, but nevertheless refers to it as the ‘rod of God’ (line 225) having previously mentioned the Hebrews’ complaint to ‘their father’s God’ (line 312).18

Finally, so far as one can tell, Demetrius the Chronographer also portrayed Moses as an MNP, for in the sole Moses miracle he narrates, the sweetening of the waters of Marah, Moses again acts on God’s instructions when he casts wood into the fountain (Praep. Evang. 9.29.15).

There are not many references to royal miracles in this literature, but there are a few. First, the Testament of Solomon attributes an almost magical ability to Solomon to control demons and employ them to carry out his will (though this ability is first of all granted to Solomon by a ring sent to him from God via the archangel Michael—T. Sol. 1.6–7). Secondly, lab 60 portrays the future king David as a kind of quasi-exorcist. Finally, Sirach 48:17–22 takes at least a step towards portraying Hezekiah as a PNP, in that he was in Sirach’s view the most noteworthy of the intercessors who prevailed upon God to smite Sennacherib’s army through his angel. To be sure these are slender pickings, but in conjunction with royal psalms such as Psalm 72, the prophecies of an ideal king such as Isa. 9:6–7; 11:1–9, and the kind of hopes expressed in Psalms of Solomon 17 and 18, there was something that could have been built on.

We have already seen that miracles could be attributed to evil figures such as Beliar or Antichrist, in passages such as Sib. Or. 2.167; 3.62; Mart. Isa. 4.1–13 and Apoc. Elij. 3.5–12. Outside this eschatological context, Mastema is credited with aiding the Egyptian magicians in the time of Moses at Jub. 48.9–11, whereas at a lower level, Aod the magician is said to have led the Israelites astray during the period of the Judges by showing them the sun at night (LAB 34.1–5). The fragmentary Jannes and Jambres tells of a pair of Egyptian court magicians who attempted to oppose Moses’ signs and wonders by replicating them. This is hardly an extensive catalogue, but it indicates that miracles could be viewed with suspicion, either as mere magic or else as full-blown demonic counterfeits.

The foregoing categories encompass most of the human miracle-workers that appear in this literature, but for the sake of completeness we should note the odd exceptions that fit none of them. Sirach 46:4–6 recalls two of the biblical miracles associated with Joshua. At LAB 27.7–11 the (biblically minor) judge Kenaz becomes a powerful PNP when his prayer for victory is answered. In 3 Maccabees 6, ‘a certain Eleazar, famous among the priests of the country’ utters a long prayer recalling God’s mighty saving acts in the past and requesting something similar in the present; he is rewarded when God despatches a pair of angels to turn Ptolemy’s elephants back upon his own troops. Aseneth prays to God for deliverance and the swords fall from her assailants’ hands and turn to ashes (Jos. Asen. 27.7–11). Finally, 2 Macc. 3:31 represents Onias the High Priest as a PNP who successfully intercedes for the life of Heliodorus.

From this survey it emerges that prophets (including Moses) form by far the largest category of human miracle-workers in Second Temple literature, and that this is attested by a wide variety of writers. In assessing the significance of the variation in the way the miracle-working of Moses and the prophets is presented, a number of points should be borne in mind: (1) The statements that God did X, that God did X through A, and that A did X are not necessarily mutually contradictory; they can be understood as focusing on different aspects of the same reality; (2) it may not always be easy to keep these three aspects in balance in literary representation, therefore a difference of expression does not necessarily imply a difference of underlying theological view; but this is not to deny that (3) different authors may well have different emphases, preferring either to ascribe all the glory to God or else to enhance the stature of the human hero through whom God works; yet (4) where, as often happens, one finds an oscillation between MNP/PNP and BNP language applied to human figures in the same text, one may well be justified in seeing the latter as an elliptical way of expressing the former.

Functions of the Miraculous

a. Healing

Although the topic of healing does occur in Second Temple literature (see, for example, the discussions on Ben Sira 38:1–15 and on healing in Philo above), stories of healing miracles are few and far between. Philo, Pseudo Philo and Jubilees have none at all, and Josephus has very few. The Wisdom of Solomon recalls a couple of healing miracles, or at least cessations of wilderness plagues, in the course of its exodus syncriseis, and Ben Sira alludes to a handful of healing or resuscitation type miracles associated with the prophets Elijah, Elijah and Isaiah. On the face of it, the clearest example of a healing miracle parallel with those in the Gospels is the healing of Tobit’s blindness in the eponymous book, though on closer examination this turns out to be more medical than miraculous. One might see a healing of sorts in Abraham’s dealings with Pharaoh in the Genesis Apocryphon (1QApGen), and it may be that God intervenes to heal the king in the Prayer of Nabonidus (4QPrNab); moreover the first of these texts mentions the laying on of hands in connexion with the cure. Both Jub. 23.14–31 and the Messianic Apocalypse (4Q521) appear to associate miraculous healing with the end-time, but on closer examination it may well be that the language is metaphorical, describing the restoration of the nation rather than individual healing miracles. Nevertheless, even if the language is only metaphorical in these cases, it may help to provide some context for making individual healing miracles an acted metaphor or symbol of eschatological salvation. The hopes for the abolition of illness at 2 Bar. 73.2 may also be relevant in this connexion.

Elsewhere the pickings become lean indeed. 2 Macc. 3:31–34 narrates the miraculous recovery from injuries of one Heliodorus at the request of the Jewish High Priest, but this is from an even more miraculous assault inflicted on him as a punishment for raiding the temple funds (3:25). Beyond that, such references to healing as there are tend to be confined to biblical characters, for example Isaiah (Sirach 48:23), Moses (Wis. 18:20–25), Elijah (Liv. Proph. 10.6; 21.7), Elisha (Liv. Proph. 22.9, 12) and Noah (Jub. 10.12). These references exhibit an awareness that miraculous healings could be associated with powerful prophets like Moses, Elijah and Elisha, but they do not indicate any great interest in healing, or any great hopes attached to the figure of the healer.

Although one might expect the notion that God is the author of healing to be common in Second Temple Judaism, there is little in this literature that actually says so. The clearest statement is perhaps that at Wis. 16:12. The praises of God in Hell. Syn. Pr. 4.24 includes the fact that his eternal power ‘raises those who are sick’. Likewise Hell. Syn. Pr. 7.6 says inter alia of God that ‘from diseases he healed’. But these texts are of limited value as evidence for Second Temple Judaism. For one thing, they come via a Christian source, although they may be Jewish synagogal prayers in origin. For another, they cannot be earlier than about 150 ce in origin.

Oblique evidence that healing was thought to be a divine prerogative is supplied by the polemic against idols in the Letter of Jeremiah. At Ep. Jer. 6:37 it is observed that ‘They cannot restore sight to a blind man; they cannot rescue a man who is in distress.’ Likewise we are told at 6:41 that ‘when they [the Chaldeans] see a dumb man, who cannot speak, they bring him and pray Bel that the man may speak, as though Bel were able to understand’. This polemic takes it for granted that any real god would be able to give sight to the blind and speech to the dumb.

Even taken together, this evidence is too meagre to draw many conclusions about which social location or variety of Judaism may have been more or less interested in healing miracles in comparison with the others. One might hazard a guess that wealthy Hellenized Jews (such as Philo) with access to the best professional medical care of the day would have less interest in healing miracles than impoverished Palestinian peasants, but the here the distinction between Palestinian and Hellenistic Judaism seems scarcely relevant: it was the Greek-speaking author of Wisdom of Solomon who was concerned to stress that healing comes directly from God and the conservative Hebrew writer Ben Sira who stressed the value of physicians. Of more significance, perhaps, is that there seems to be slightly more interest in healing miracles in literature from the Enochic-Qumran traditions. This tendency turns out to be even more pronounced in the case of exorcism.

b. Exorcism

That belief in demons was common among late Second Temple Jews is uncontentious, and is well attested in the literature under discussion. But although demons are widely attested in this literature, they are far from ubiquitous. Many texts do not mention them at all. In 4 Ezra, for example, the blame for humanity’s condition is placed firmly on humanity: demons are not blamed, and neither is their ultimate destruction in view. Unlike Jubilees, Pseudo-Philo feels no need to introduce demons into his re-telling of the biblical narrative (with one exception). Demons are similarly absent, for example, from Philo, Artapanus, the Lives of the Prophets, the apocryphal additions to Daniel, Ben Sira, the Wisdom of Solomon, the Letter of Jeremiah, Judith, 1, 2, 3 and, 4 Maccabees, the Psalms of Solomon, and Joseph and Aseneth. This negative catalogue could be continued, but the point should now be clear. Although belief in demons is widely attested in this literature, only a minority of the surviving texts actually attest it.

Actual stories of exorcisms are even rarer than healing stories. Perhaps the best example for the purposes of the present study is that provided by Josephus at Ant. 8.46–49 (Eleazar). This does appear to provide an account of a man who was possessed in the narrow sense and from whom the demon was driven out. Elsewhere it is less clear that possession in this strict sense is ever in view. Thus there are exorcism stories in Tobit (where Sarah’s husbands, not Sarah herself, seem to have been the victims), the Genesis Apocryphon (where the evil spirit has been protecting another Sarah from her would-be husband), and LAB 60 (which is more interested in David’s psalm than Saul’s affliction), but none of these is particularly similar to the stories in the Gospels or Acts. Even less similar is the story of Noah’s prayer against the demons at Jub. 10.1–9 although, unlike the other examples, it does have more of an ‘eschatological’ effect in that it results in nine-tenths of the demons being bound.

The Qumran literature contains a number of apotropaic texts in some way related to exorcism, but I have argued that few of these are exorcistic in the strict sense; most seem to be concerned with prevention rather than cure. Moreover, it is not clear that these texts always envisage demonic attack taking the form of possession in any strict sense (either the narrow sense of displacing the victim’s personality, or the broader sense of entering the victim’s body to cause illness). In the light of the Qumran two-spirits doctrine it sometimes seems possible to interpret them as protection against demonic temptation. Indeed the ills attributed to evil spirits in the Enochic tradition (especially 1 Enoch and Jubilees) range far more widely than possession, while possession in the narrow sense is scarcely ever mentioned.

For example, in 1 Enoch 8 Azaz’el and his companions teach human beings such dangerous arts as making weapons, the use of cosmetics and other ornamentation, incantations and astrology. At 1 En. 15.8–12 the Watchers are to ‘rise up against the children of the people and against the women’ and ’cause sorrow’, but their principal sin was to broadcast forbidden mysteries to women so that ‘women and men multiply evil deeds upon the earth’ (16:3). At 40.7 a voice expels the demons from heaven so that they can no longer come ‘to the Lord of Spirits in order to accuse those who dwell upon the earth’. The misdeeds of the fallen angels are spelled out in some detail at 69.4–15; they include misleading people, teaching humankind how to make instruments of death, and imparting the art of writing. In 1 Enoch 86 the wicked angels appear in the guise of falling stars, who came down to mate with human females, resulting in havoc and slaughter upon the earth (through the violent activities of the giants who are the offspring of the union between and angels and women). Although 1 Enoch is clearly a composite book, the general picture of the demonic that runs through it sees demons as primarily responsible for instigating human sin, rather than for invading individual human personalities.

The later Testament of Solomon (containing clear Christian influences) is full of demons. Throughout the work they are obliged to assist Solomon in the building of the temple, and are therefore conceived as physical beings who can manipulate physical objects (their physical appearances are also described). The demons questioned by the king admit to such diabolical activities as creating divisions, starting fires and whirlwinds, making households dysfunctional, making men fight in battle, leading people into error, raising up tyrants, spreading wickedness, preventing recovery from illness, strangling new-born infants, copulating with women, marring the beauty of virgins, drowning people, causing sea-sickness, killing people with a sword, and causing a variety of physical ailments. But only at one place is demon-possession explicitly mentioned, where the spirit of a deceased giant announces that if he is unable to kill someone, ‘I cause him to be possessed by a demon and to gnaw his own flesh to pieces and the saliva of his jowls to flow down’ (17.3). The lion-shaped demon of chapter 11 commands a legion of demons whom, he predicts, will be tormented by being driven into the water at the cliff by Emmanouel (a clear reference to the story of the Gerasene demoniac), but it is not actually said that the demons have just been ejected from a human victim into a herd of pigs.

Thus where demons or evil spirits are mentioned, they are held responsible for all sorts of mayhem, but they are hardly ever represented as invading an individual human personality and taking control. It is thus hardly surprising that, although the defeat of the evil powers is described in a number of ways, exorcism is not among them. It is instead variously portrayed in terms of binding (1 En. 10.4; T. Levi 18.12), chaining (1 En. 69.28), throwing into eternal fire (T. Jud. 25.3), expulsion from heaven (1 En. 40.7), casting into an abyss (1 En. 54.5; 88.1), imprisonment (1 En. 10.13), or simply having an end (T. Mos. 10.1).

Sanders is therefore correct in saying that ‘the view that the Messiah (or his herald) was expected in Judaism to overcome the demonic world and to demonstrate this victory miraculously by exorcisms … is hard to find’. Yet there may be more to it than that; although the view as Sanders states it is certainly hard to find, it may represent the confluence of several streams in Judaism that can be found rather more readily. It is surely of some interest that the story that most closely resembles the Gospel exorcism stories most closely comes from Josephus, who otherwise shows little interest in the demonic and attaches no eschatological significance to exorcism, while a stream of literature from the Enochic-Qumran tradition is greatly exercised with the demonic and its ultimate defeat but relates virtually no exorcism stories. The only place in Jewish literature where these two streams almost touch is in Tobit, where the expulsion and binding of Asmodeus perhaps assimilates him to the fallen angels that are bound in 1 Enoch and Jubilees (perhaps because for the author of Tobit the overcoming of the afflictions of the various Jews in his story is paradigmatic of the hoped-for return from exile of the people as whole).

The material on healing and exorcism has not yet been exhausted. Chapter 12 will examine the evidence for Jewish exorcists, and Chapter 13 will outline the contribution of the social sciences to understanding folk-healing and spirit possession. To anticipate those discussions, perhaps the best way of understanding the context of Jesus’ healings and exorcisms is at the creative confluence of three streams within Judaism: the widespread traditions about prophets like Elijah and Elisha, the Enochic-Qumran traditions concerned with the eschatological defeat of demonic powers, and popular folk-religion.

c. Nature Miracles

It must next be considered whether there are any parallels to the Gospel ‘nature miracles’ in this literature, that is any references to walking on water, controlling a storm at sea, miraculous feeding and the like.

The only (non-Christian) reference to anyone walking on the sea is a sarcastic sneer at Antiochus Epiphanes, at 2 Macc. 5:21. More relevantly, 1 En. 101.4–7 testifies to the belief that storms at sea were under God’s control. In common with Ps. 104:6–7 it provides an example of Yahweh rebuking the elements (cf. Mk 4:39). It is, however, in allusions to the miracle at the Red Sea that the greatest number of water-crossing miracles in this literature are found. But in the Apocrypha and Pseudepigrapha (as in Philo and Josephus) references to this miracle and its imitators generally emphasize the piling up of the waters, the emergence of dry land, or the cessation of the flow of a stream (as in the account of Joshua’s crossing of the Jordan).

Thus, for example, 4 Ezra 13:39–45 envisages an eschatological exodus for the ten lost tribes, in the course of which, ‘the Most High performed signs for them, and stopped the channels of the river until they had passed over’ (13:44). Similarly, when in The Lives of the Prophets Ezekiel is credited with performing a miracle by the Chebar, the echoes of the exodus story are unmistakable: ‘And he made the water stop so that they might escape by getting to the other side. And those of the enemies who dared to pursue were drowned’ (3.8–9). The same book insists that both Elijah and Elisha struck the Jordan and divided it, and crossed over with dry feet (21.14; 22.5).

These features are also present when the exodus itself is narrated. Thus, for example, Jub. 48:13–14 recalls:

And the Lord brought them out through the midst of the sea as through dry land. And all of the people whom he brought out to pursue after Israel the Lord our God threw into the middle of the sea into the depths of the abyss beneath the children of Israel. Just as the men of Egypt cast their sons into the river he avenged one million.

With this one might compare Pseudo-Philo’s account (LAB 10.5–6) and those of Artapanus (Art. 27.36–37), and Ezekiel the Tragedian (Ezek. Trag. 220–42). These passages suggest that certain elements formed standard features of any narration of the Red Sea crossing: Moses’ rod, the parting of the waters and the Israelites’ passage as if walking on dry land, their desperate plight in the face of the pursuing Egyptians and the Egyptians’ destruction in the waters of the sea.

At least one of these features is normally present even when the exodus is only alluded to. Thus the Synagogal prayer possibly preserved at Hell. Syn. Pr. 5.12 says that in the course of rescuing Israel from Egypt, ‘you led them through (the) sea as through dry land’. Likewise at Hell. Syn. Pr. 12.73 the same event is described in the words, ‘you carried the Israelites across a sea, dividing it; you destroyed the pursuing Egyptians under water’. According to Sirach, ‘At his [God’s] word the waters stood in a heap’ (Sirach 39:17). The book of Wisdom refers to the event more than once. At Wis. 10:18–19 it is said that wisdom ‘brought them over the Red Sea, and led them through deep waters; but she drowned their enemies, and cast them up from the depth of the sea’. Wisdom 18:5 again alludes to the drowning of the Egyptians, while Wis. 19:7–8 once again emphasizes that the Israelites passed through on dry land.

These ‘standard’ features are largely absent from the sea miracles of the Gospels (as will be discussed further in Chapter 14 below). This is not to deny that exodus allusions may be found at the redactional level of the Gospel narratives, not least in the juxtaposition of a feeding story with a sea-crossing story in Mark 6, John 6 and Matthew 14. The point is the absence of characteristic exodus features when the stories are taken in isolation. As I shall now go on to argue, this is equally the case with provision miracles such as the feeding stories.

LAB’s account of the miraculous provision in the wilderness is brief: ‘for forty years he rained down on for them bread from heaven and brought quail to them from the sea’ (LAB 10.7; cf. Ps. 78:24). Philo also envisages the manna as food rained down from heaven, and dwells at some length on its special properties (Vit. Mos. 1.196–206; 2.266). Hellenistic Synagogal Prayers 12.75 notes that God ‘rained the manna out of heaven, a mother-of-quail nourishment out of the air’, while according to Artapanus‘s description, ‘God rained for them meal like millet, very similar in color to snow’ (Art. 27.37). Wisdom 15:18–16:4 contrasts the quails provided to the Israelites with the animals used to punish the Egyptians (who worship animals), but does not dwell on how the provision was made. Likewise Wis. 16:15–29 contrasts the hail and lightning that destroyed the Egyptians’ crops with the manna given to the Israelites.

Wisdom is not explicit on the matter, but all these other texts envisage the manna as food rained down from heaven. With this we might compare the expectation of 2 Bar. 29.7–8 that when the Anointed One begins to be revealed:

winds will go out in front of me every morning to bring the fragrance of aromatic fruits and clouds at the end of the day to distill the dew of health. And it will happen at that time that the treasury of manna will come down again from on high, and they will eat of it in those years because these are they who will have arrived at the consummation of time.

In contrast, the bread in the Gospel stories is not said to have come down from heaven. It is only the Fourth Gospel that makes this connexion, and then not in the story itself, but in the discourse that follows it. The Gospel stories are also not especially similar to the occasional depictions of the so-called ‘messianic banquet’. As it so happens, the last text referred to, 2 Baruch 29, contains one of the clearest descriptions of such a feast:

And it will happen that when all that which should come to pass in these parts has been accomplished, the Anointed One will begin to be revealed. And Behemoth will reveal itself from its place, and Leviathan will come from the sea, the two great monsters which I created on the fifth day of creation and which I shall have kept until that time. And they will be nourishment for all who are left. The earth will also yield fruits ten thousandfold. And on one vine will be a thousand branches, and one branch will produce a thousand clusters, and one cluster will produce a thousand grapes, and one grape will produce a cor of wine. And those who are hungry will enjoy themselves and they will, moreover, see marvels every day (2 Bar. 29.3–6).

This passage then continues with the section about the manna. The whole passage is notable, then, not only for its description of superabundance in the Messianic age, but for its association with daily marvels in general and a repetition of the manna miracle in particular. Compared with this the feeding of the five thousand looks remarkably modest. Most scholars date 2 Baruch to the end of the first century ce, on the grounds that it reflects both the destruction of Jerusalem in 70 ce and that it appears to be related to 4 Ezra. The traditions contained in 2 Baruch may well be earlier, but we cannot be certain that the relevant ideas were current at the time of Christ or even of Mark.

Another account of miraculous plenty at the eschaton, which may also contain an oblique reference to a repetition of the manna miracle, occurs at Sib. Or. 3.741–51:

When indeed this fated day also reaches its consummation

and the judgment of immortal God comes upon mortals,

a great judgment and dominion will come upon men.

For the all-bearing earth will give the most excellent unlimited fruit

to mortals, of grain, wine and oil

and a delightful drink of sweet honey from heaven,

trees, fruit of the top branches, and rich flocks

and herds and lambs of sheep and kids of goats.

And it will break forth sweet fountains of white milk.

The cities will be full of good things and the fields will

be rich. There will be no sword on earth or the din of battle.

This passage goes on to promise a cessation of natural disasters such as famine, hail and drought, together with universal peace, and a common law for all humanity imposed by the one true God. Many of these ideas may well go back to Old Testament passages such as Isa. 2:2–4. It would not be unreasonable to describe this as one strand of expectation concerning a future ‘kingdom of God’.

A third passage in this literature that describes an age of plenty occurs at 1 En. 10.17–19, following a description of the judgment that is to fall on the Watchers and their brood. This, however, is a description of wondrous fertility rather than of either a ‘Messianic banquet’ or a duplication of the manna miracle.

There are one or two other miracles of provision scattered across this literature. Liv. Proph. 1.2–3 tells of a miraculous provision of water for the dying Isaiah, and previously for the besieged Hezekiah. At Liv. Proph. 3.10 Ezekiel is credited with a miraculous catch of fish, effected through prayer, to feed his starving compatriots. If this vaguely resembles the miraculous catches of fish in Luke and John, the story of Tobias’ catching a fish at Tob. 6:2–5 is an even more distant relative of Matthew’s account of the coin in the fish’s mouth. At Liv. Proph. 12.7 Habakkuk is credited with miraculously travelling to Babylon to feed Daniel. This same miracle is described in more detail in Bel and the Dragon 33–39. Liv. Proph. 21 summarizes the provision miracles attributed to the biblical Elijah: rain comes when he prays, the widow of Zarephath’s oil does not run out and her jar does not fail. Lives of the Prophets 22 similarly summarizes Elisha’s miracles, which includes another miraculous provision of oil but does not include the feeding of a 100 men from 20 loaves.

With the exception of the miracles taken over from the Old Testament, none of these stories significantly resemble the Gospel feeding accounts. Overall, then, while this literature expresses some expectations of an age of plenty, a repetition of the manna miracle, and a Messianic banquet, it cannot be said that these ideas are widely attested, or that the Gospel feeding stories obviously originated from them.

d. Miracles and Eschatology

The word ‘eschatology’ is often used loosely to refer to any kind of future expectation of a decisive intervention in the course of events. Here, however, its use will be restricted to the narrower sense of expectations about the end, in the sense of the consummation of history when all the conflicts of the present age are finally resolved and God’s purposes are fully realized.

Accounts of the end in this literature frequently contain descriptions of bizarre events and cosmic disturbances, which one might term ‘miraculous’. Sibylline Oracles 3.796–808, for example, contains a description of signs of the end that is reminiscent both of the portents mentioned in Book 6 of the Jewish War and of some of the conventional language in Mark 13 and elsewhere. Sibylline Oracles 3.54 threatens a ‘fiery cataract from heaven’ directed particularly against Rome. The ending of the whole world through conflagration is depicted in more detail in Sib. Or. 4.171–78. The notion that the end will be accompanied by general disorders on a cosmic scale is also common in Jewish apocalyptic. 4 Ezra 5:1–13 describes such an upheaval, and in vv. 4 and 5 it comes very close to the language of Sib. Or. 3.796–808. 4 Ezra 6:18–24 likewise describes a sequence of bizarre happenings at the end.

The theme of general calamity, destruction and terror is also found at 1 En. 102.1–3. 2 Baruch 27 describes the social and natural disorders of the 12 periods leading up to the end. The language about events preceding the judgment at T. Levi 4.1 and T. Mos. 10.3–6 is again reminiscent of 4 Ezra and the Sibylline Oracles. Here it will suffice to quote one passage, again from 4 Ezra, which explicitly describes the eschatological confusion as mighty works, signs and wonders:

So when there shall appear in the world earthquakes, tumult of peoples, intrigues of nations, wavering of leaders, confusion of princes, then you will know that it was of these that the Most High spoke from the days that were of old, from the beginning. For just as with everything that has occurred in the world, the beginning is evident, and the end manifest; so also are the times of the Most High: the beginnings are manifest in wonders and mighty works, and the end in requital and in signs (4 Ezra 9:3–6).

Although 4 Ezra describes these things as mighty works, wonders and signs they seem a far cry from the miracle stories of the Gospels, and more akin to eschatological discourses such as Mark 13. It is unclear how far this language is to be taken literally, how far it is merely conventional, or how far it is to be understood as symbolic (perhaps of a political collapse).

The questions of eschatological feeding and the defeat of demonic powers have been dealt with above. This still leaves a few remarks to be made about ‘eschatological’ healing.

Isa. 35:5–6 is often cited as expressing expectations that Jesus’ healings could be seen as meeting. There are, however, few indications that this text was read as expressing eschatological expectations in this literature. The only clear example is Sib. Or. 8.205–207:

There will be a resurrection of the dead

and most swift racing of the lame, and the deaf will hear

and blind will see, those who cannot speak will speak.

Unfortunately, the authorship of this passage is not clear: it could well be Christian. It would thus be perilous to cite this passage as evidence for a first-century bce–ce Jewish understanding of Isaiah 35. Beyond this passage (and Apoc. Elij. 3.5–10) it is hard to find anything in this literature that specifically associates healing with the eschaton. I have already argued that in passages such as Jub. 23.23–31 and the Messianic Apocalypse, healing may be a metaphor for national restoration and general wellbeing. The only other passage in this literature that makes an apparent connexion between healing and the end time is in the Life of Adam and Eve (41–42) and the parallel passage in Apoc. Mos. 13. Adam is dying, and sends Eve and Seth to Paradise to request oil from the tree of mercy to alleviate his pain. But after several hours of entreaty the angel Michael announces, ‘Truly I say to you that you are by no means able to take from it, except in the last days’ (LAE 42.1). An expansion of this speech that appears in a minority of the Greek manuscripts of Apoc. Mos. 13.3–5 adds that at that time all flesh shall be raised, and there shall be no more sinners before God because he shall have replaced their evil heart with an understanding one. The reservation of the ‘oil of mercy’ until the last days thus suggests that the relief of pain will be part of the eschatological blessings that reverse the dire consequences of the fall. Only very loosely does this associate healing with the eschaton.

What this evidence indicates is that there was no automatic connexion between healing and eschatology in Jewish thought, but that it would not be beyond the bounds of possibility to associate the two. This perhaps strengthens the idea that Jesus made creative use of individual healings as an acted metaphor (or parable) of the coming of the Kingdom of God.

e. Miracles and Salvation History

Miracles might accompany a decisive turn of events without being strictly eschatological. Josephus’s theory of divine providence perhaps falls into this category, as does Pseudo-Philo’s insistence on the divine control of events. The most notable examples of this kind of miracle in Jewish tradition are provided by the accounts of the exodus. We have already examined the exodus-type traditions in this literature for resemblances to the Gospel miracles; we must now examine these and similar traditions to see how they were employed in future hopes of a salvation-historical kind.

Perhaps the most explicit use of the exodus traditions in this way occurs in the Wisdom of Solomon, as we have seen.

The prayer for divine aid at Sirach 36:1–17 reflects a similar hope:

Show signs anew, and work further wonders;

make thy hand and thy right arm glorious.

Rouse thy anger and pour out thy wrath;

destroy the adversary and wipe out the enemy.

Hasten the day, and remember the appointed time,

and let people recount thy mighty deeds (Sirach 36:6–8).

To be sure, there is no explicit reference to the exodus here, but the language is similar to that in which the exodus events were frequently recalled (e.g. Bar. 2:11; Deut. 4:34; 5:15; 7:19; 9:29; 11:2; 26:8; 2 Kgs 17:36; Ps. 136:12; Jer. 32:21) though it could also be used when God was expected to act in a broadly analogous fashion (1 Kgs 8:42; 2 Chron. 6:32; Ezek. 20:33–34). Whatever the specific reference, this is clearly a plea that God should come to the aid of his people as he had in times past.

One miracle of deliverance recalled several times in this literature is the slaughter of 185,000 Assyrians by the angel of the Lord. Although 1 Maccabees contains no miracles, it contains a prayer in which Judas asks that God might help him defeat the enemy (Nicanor) as when the angel slew 185,000 Assyrians (1 Macc. 7:41). In 2 Maccabees Judas alludes to this event both in a prayer (2 Macc. 15:22) and in a speech to encourage his troops (8:19). Sirach’s ‘Praise of Famous Men’ regards this event as one of the more notable of the reign of the godly king Hezekiah (Sirach 48:21).

Other spectacular divine interventions are also recalled at times of crisis. 3 Maccabees 2:1–20 gives a lengthy prayer by the high priest Simon on the occasion of Ptolemy’s attempt to enter the holy of holies: Simon and all the people regard this with such horror that Simon reminds God not only of the exodus miracles, but also of the destruction of Sodom and his love of Israel, whereupon God responds by striking Ptolemy with paralysis. Mattathias’s testamentary discourse at 1 Macc. 2:49–68 encourages his sons to remember the deeds of their forebears, among which he includes Elijah’s ascent into heaven, Daniel’s deliverance from the lions, and his three friends’ salvation from the furnace. But Mattathias also cites many non-miraculous examples, and his point is that God will supply strength to those who put their trust in him, rather than miraculous aid.

Thus, where miracles are associated with salvation history in this literature, their function is to punish enemies and gain deliverance for the faithful, normally conceived of as military victory or deliverance from foreign domination.

f. Punishment Miracles

As in Josephus, Philo and, indeed, the Old Testament, many of the punishment miracles in this literature are the shadow side of the miracles of national deliverance (as we have just seen). But there are also a number of purely punitive miracles, as I shall now illustrate by means of a few examples.

Some of the punitive miracles are simply further accounts of biblical stories we have already encountered elsewhere, such as the destruction of those who took part in Korah’s rebellion (LAB 16; Sirach 45:18–19). Others are additions to the biblical narrative, such as the fire that destroyed the men who worshipped Baal at Jair’s altar (LAB 38) and the miraculous punishment of Micah and his mother for their idolatry (LAB 44.9–10; 47.12). These examples show people punished for idolatry or rebellion, two forms of opposition to God. At Jan. Jam. 26a Jannes is said to oppose Moses’ signs and wonders by replicating them; he is punished by being struck with an ulcer. At 3 Macc. 2:21, in response to the prayer of Simon the high priest, God punishes Ptolemy for trying to enter the Holy of Holies. Both these episodes may also be said to involve opposition to God in one way or another; Jannes tries to obscure the divine origin of the miracles wrought through Moses, and Ptolemy proposes to profane the sanctuary, something that God is said to regard as ‘insolence and audacity’ (3 Macc. 2:21). In similar vein, Alcimus is paralysed and then dies in agony for ordering the sanctuary wall torn down (1 Macc. 9:54–57), though here the offence is said to be against ‘the work of the prophets’ rather than God directly. At 2 Macc. 9:5 God similarly punishes Antiochus for his arrogance when he proposes to vent his anger on the Jews of Jerusalem; he is ‘seized with a pain in his bowels for which there was no relief’. Tfhe modern reader may, however, feel that the last four of these punishments, involving illnesses inflicted on Jannes, Ptolemy, Alcimus and Antiochus, were not all that miraculous, although they do illustrate how illness could be seen as a divine punishment.

Most of the other punishment miracles in this literature have already been considered under other headings. Many of the ones considered here also averted some threat directed at Israel (for example, Antiochus was prevented from attacking Jerusalem, Micah was stopped from leading Israel astray with his idolatry, and Jannes was hindered in his opposition to the man who was about to lead Israel’s exodus). Apart from eschatological punishments, which are also mentioned in the New Testament, it cannot thus be said that this literature adds greatly to the store of miracles imposed purely for the sake of punishment; in most cases some kind of protection or deliverance for God’s people is in view, even if it is sometimes only in the background.

g. Evidential and Accreditation Miracles

In Josephus miracles accrediting God’s prophet formed an important class, and Philo stressed the evidential value of miracles. These themes do not, however, emerge strongly in the Apocrypha and Pseudepigrapha. Liber Antiquitatum Biblicarum provides a spectacular account of the sign given to Gideon, in which water becomes half blood and half fire (LAB 35.6–7); Jub. 48.4 and Ezek. Trag. 120–31 refer to Moses’ signs and wonders and, as we have just seen, Jan. Jam. 26a refers to Jannes’s attempt to undermine their evidential value by replicating them; Sirach 48:23 alludes to the sign of the reversing sun given to Hezekiah. In this literature, signs and wonders are just as likely to be used to mislead (Jan. Jam. 26a; Beliar at Syb. Or. 2.167; 3.63–70 and Mart. Isa. 4.1–13; Aod the magician at LAB 34). This is not a large catalogue from so large a collection of literature. Insofar as one may draw any conclusions from it, one might say that this literature is just as alive to the epistemological ambiguity of miracle as Josephus appears to be, but shows rather less interest in the use of miracles as authenticating signs. Thus, apart from Josephus and the request for a sign directed at Jesus in the Gospels, there is actually little hard evidence that a prophet in Jesus’ day would be expected to produce an authenticating sign; and in Chapter 11 it will be argued that the signs promised by Josephus’s ‘sign-prophets’ were not primarily for the purpose of accreditation.


This chapter has attempted to draw together the threads of the previous seven. The purpose has not been to produce a synthetic account of the Jewish view of miracle exhibited in this literature. Rather the aim has been to identify what ideas about miracle tend to recur and which are rare, and whether any ideas are more characteristic of one variety of Judaism than another. It turns out that there is relatively little interest in miracles of healing anywhere and far more interest in the spectacular miracles of national deliverance. Again a wide variety of texts attest that Moses and the prophets are the human figures most commonly associated with miracle working. These trends thus point to features of miracle belief that were clearly widespread amongst a variety of Jews at the time. We have further noted how the great miracles of the exodus period (especially the crossing of the Red Sea and the manna in the wilderness) tend to be narrated in ways that make them quite distinct from the superficially similar miracles in the Gospels (with which they are often compared).

The Apocrypha and Pseudepigrapha go beyond Philo and Josephus in two ways particularly relevant to the present study. First, some of these texts express specific eschatological expectations, but where miracles feature in these they are usually quite unlike the miracles of Jesus. In what I have termed the Enochic-Qumran traditions, however, there is an interest in healing and the ultimate defeat of demonic powers. This may well form one strand that was subsequently taken up in the Jesus tradition. Secondly, a few texts go beyond the epistemological ambiguity recognized by Josephus and see miracle-working as potentially leading people astray, a theme to which we shall return in connexion with spirit possession and false prophecy in Chapter 13.

This survey of late Second Temple Jewish literature thus far has confirmed Harvey’s contention that Jesus appears unique in the number and type of healings and exorcisms he performs. No other broadly contemporary figure is portrayed thus, and miracles of healing are not prominent among the signs and wonders this literature generally celebrates. This survey has also broadly confirmed Kahl’s thesis, at least in relation to Jewish literature, that Jesus is unique in being presented as an immanent BNP to whom a series of miracles is attributed, although Artapanus‘s Moses comes close. In most of this literature the only BNP is God, and this is particularly stressed in the extensive writings of Philo and Josephus. Yet we have also seen, occasionally even in Philo, more often in the Apocrypha and Pseudepigrapha, some instances where BNP language can be used of a human miracle-worker closely associated with God, so that some kind of synergy may be in view.

This completes the survey of Second Temple Literature, but it does not complete the investigation into the Jewish context of Jesus’ miracles. The next three chapters will look at the evidence for Jesus’ possible miracle-working contemporaries, starting with the charismatic holy men proposed by Geza Vermes.

Chapter 10

Charismatic Holy Men


One often meets the view that Jesus’ miracles connect him with a substantial class of contemporary Palestinian charismatic holy men. The two people taken to exemplify this class are Ḥoni the Circle-Drawer and Ḥanina ben Dosa. This chapter will investigate whether the traditions about these men support the picture built on them.

The most elaborate miracle stories about Ḥoni and Ḥanina occur in the Babylonian Talmud (henceforth BT), the final redaction of which can be dated no earlier than the fifth century ce, if not the sixth or seventh. The material contained in BT may be much earlier than the date of its redaction, but (as we shall see) the tendency of BT is to turn Ḥoni and Ḥanina into respectable rabbis. The suspicion must be that this transformation is comparatively late, and that it tells us more about Amoraic perceptions of miracle-working rabbis than about anything that happened in the first century.

It is conceivable that BT preserved earlier traditions about figures barely mentioned in Tannaitic literature. This is not simply because the Mishnah and Tosefta are more concerned with halakah than with narratives. Morton Smith has demonstrated that the Tannaim told virtually no stories about miracles performed by other Tannaim. This may mean that few such stories existed, but it may instead mean that the Tannaitic rabbis were wary of charismatic challenges to their authority, and so tended to suppress stories about prophets, magicians and miracle-workers.3 In principle, then, it could be argued that historical traditions about Second Temple figures were transmitted by routes that are largely invisible in the surviving Tannaitic literature, only to surface in Amoraic literature.

Nevertheless, this possibility is prima facie unlikely. In his study of the transmission of folk traditions from the Second Temple to the talmudic period, Eli Yassif found first that only a small stock of folk-traditions survived, secondly that those that survived tended to be ones that could be given a new function, thirdly that ‘In the process of transmission from period to period, traditions underwent major alterations in form, theme and function, changes which can be explained against the background of the new historical and ideological setting in which the stories were re-told’, and fourthly that the survival of such folk-traditions was largely dependent on the continuation of the appropriate written sources. The probability that reliable historical information about such Second Temple figures as Ḥoni and Ḥanina was conveyed to the redactors of BT through several centuries of oral tradition is therefore remote.

The most thorough case for taking Ḥoni and Ḥanina as parallels to Jesus has been made by Geza Vermes. This chapter will accordingly focus on his views; its main concerns are to determine what can be known about the historical Ḥoni and the historical Ḥanina, whether this provides an adequate basis for postulating the existence of a class of charismatic holy men in late Second Temple Judaism, and to what extent these figures are comparable to Jesus.

Ḥoni the Circle-Drawer

Vermes introduces Ḥoni to his readers thus:

One of the prime characteristics of the ancient Ḥasidim or Devout is that their prayer was believed to be all-powerful, capable of performing miracles. The best known of these charismatics, though perhaps not the most important from the point of view of New Testament study, is a first-century bc saint, called Ḥoni the Circle-Drawer by the rabbis, and Onias the Righteous by Josephus.

Almost every assertion in this paragraph could be challenged. The Mishnah does not call Ḥoni a Ḥasid, neither does Josephus refer to anyone as ‘Onias the Righteous’, although he does describe his rain-making Onias as ͂δίκαιος ἀνήρ καὶ θεοφιλής (Ant. 14.22). It is not even certain that Josephus and the Mishnah are talking about the same person, although it seems likely. More worrying is the transition from ‘devout’ to ‘charismatic’, and the unsupported suggestion that Ḥoni was the best known of a sizeable class of such evout-charismatics. Whether the Ḥoni tradition presents Ḥoni’s prayer as ‘all-powerful, capable of performing miracles’ is also open to question.

The only Tannaitic account of Ḥoni the Circle-Drawer occurs in the Mishnah, at Ta’an. 3.8:

They sound the shofar because of any public distress—may it never befall! But not because of too great abundance of rain. Once they said to [Ḥoni] the Circle-maker,’ Pray that rain may fall’. He answered,’ Go out and bring in the Passover ovens that they be not softened’. He prayed, but the rain did not fall. What did he do? He drew a circle and stood within it and said before God, ‘O Lord of the world, thy children have turned their faces to me, for that I am like a son of the house before thee. I swear by thy great name that I will not stir hence until thou have pity on thy children’. Rain began falling drop by drop. He said, ‘Not for such rain have I prayed, but for rain that will fill the cisterns, pits and caverns’. It began to rain with violence. He said, ‘Not for such rain have I prayed, but for rain of goodwill, blessing and graciousness’. Then it rained in moderation [and continued] until the Israelites went up from Jerusalem to the Temple Mount because of the rain. They went to him and said, ‘Like as thou didst pray for the rain to come, so pray that it may go away!’ He replied, ‘Go and see if the Stone of the Strayers has disappeared!’ Simeon b. Shetaḥ sent to him [saying], ‘Hadst thou not been [Ḥoni] I had pronounced a ban against thee! But what shall I do to thee?—thou importunest God and he performeth thy will, like a son that importuneth his father and he performeth his will; and of thee Scripture saith, Let thy father and thy mother be glad, and let her that bare thee rejoice’.

This passage raises a number of questions. First, what is it doing in the Mishnah? Second, what does it wish to convey about Ḥoni? And third, can it be used as a source of historical information about a figure who lived roughly three hundred years before?

The occurrence of this story at this particular point in the Mishnah seems odd. The Mishnah tractate Ta’anith is principally concerned with fast-days, and the part of it in which the Ḥoni story appears with occasions on which the shofar should be blown (to announce a fast or public disaster). Ta ‘an 3.8 begins by stating that the shofar is not sounded because of too much rain. One would then expect either an explanation of why too much rain is not considered a disaster, or else an account of rabbinic precedents for not sounding the shofar in cases of excessive rain. What the Mishnah actually narrates is a story about too much rain.

As W. Scott Green notes, the similar story in the Tosefta at t. Ta ‘an. 2.13 would fit the Mishnah context much better:

A.    M’ŚH B: To a certain pious man did they say, ‘Pray, so it will rain’.

B.    He prayed and it rained.

C.    They said to him, ‘Just as you have prayed that it would rain, now pray so the rain will go away’.

D.    He said to them, ‘Go and see if a man is standing on Keren Ofel [a high rock] and splashing his foot in the Qidron Brook. [Then] we shall pray that the rain will stop [cf M. Ta. 3.8]

E.    ‘Truly it is certain that the Omnipresent will never again bring a flood to the world.

F.    ‘for it is said, There will never again be a flood (Gen 9:11).

G.    ‘And it says, For this is like the days of Noah to me: as I swore that the waters of Noah should no more go over the earth, so I have sworn that I will not be angry with you and will not rebuke you (Is. 54:9).

This story would explain why too much rainfall is not to be treated as a disaster: God has promised that he will never again destroy the world by flood, so that to sound the shofar for too much rain would be to show lack of trust in God’s promise. The anonymous Ḥasid’s comment at D is presumably ironic: the rain is never going to reach the top of Keren Ofel; only if it does will it be necessary to pray for the rain to stop’.

Green argues that the Mishnaic account of Ḥoni is a redactional combination of the Tosefta’s account of the anonymous Ḥasid with a tradition that represented Ḥoni as a magician. In his opinion it is included in order to rabbinize Ḥoni. Despite the rabbinic hostility to miracle-workers, the rabbis wished to show that their brand of religion had fully taken over the functions of the temple, which included ensuring regular rainfall for the land.15 The Mishnah accordingly attempts to transform a popular rainmaker into a proto-rabbi.

A more plausible reading of the Mishnah’s Ḥoni story is that proposed by Jacob Neusner. As Neusner points out, the story does not portray Ḥoni as a successful rain-maker, but rather pokes fun at him. When asked to pray for rain, Ḥoni boastfully orders the (clay) Passover ovens to be taken inside, so confident is he of his abilities; but his prayer proves fruitless. He then stands inside his circle and demands special consideration as the child of the family. God responds by sending drizzle. Ḥoni complains that this is not what he wanted, so God turns the drizzle into a torrent. Ḥoni complains that this is still not the right kind of rain, so God now sends the right kind of rain—but too much of it, so that the people ask Ḥoni to make it stop. Ḥoni is not reported as achieving this feat, he merely attempts another boast: ‘Go and see if a certain stone is under water’, implying that he will oblige their request if and when it becomes submerged. Simeon b. Shetaḥ is then brought in to provide a rabbinical comment on Ḥoni’s activities.

Neusner suggests that the purpose of this story is to make miracle-working appear ‘undignified’. Although Ḥoni prevails upon God to send rain, he cannot control it, and it is the rabbi with his scriptural quotation who is given the last word. The Mishnah tells no stories of second-century miracle-workers, but rabbis of the third and even more the fourth centuries are frequently portrayed as miracle-workers. Neusner accordingly suggests that this polemical story about Ḥoni may owe its origin to a period where there was tension within the rabbinic movement itself, as it began to move towards the position that this story opposes.

Green and Neusner are agreed that m. Ta ‘an 3.8 cannot be used as a straightforward source about the historical Ḥoni. This view is surely correct. Either the account is purely legendary or, it is a polemical attempt to portray Ḥoni as a bumbling magician or precocious PNP. The most this indicates is that Ḥoni was reputed to be effective at interceding for rain, otherwise the rabbis would not have chosen him as a butt of their polemic. It is conceivable that the anonymous Ḥasid of t. Ta ‘an 3.20 was also Ḥoni, but this is by no means certain, and even then the story will have been developed over the course of three centuries or so. The maximum historical kernel of these Tannaitic traditions is thus that there was once a first-century bce character called Ḥoni who famously prayed for rain during a drought, following which it rained. This is no more than Josephus’s account of Onias states, as we shall now see:

Now there was a certain Onias, who, being a righteous man and dear to God, had once in a rainless period prayed to God to end the drought, and God had heard his prayer and sent rain; this man hid himself when he saw that the civil war continued to rage, but he was taken to the camp of the Jews and was asked to place a curse on Aristobulus and his fellow-rebels, just as he had, by his prayers, put an end to the rainless period. But when in spite of his refusals and excuses he was forced to speak by the mob, he stood up in their midst and said, ‘O God, king of the universe, since these men standing beside me are Thy people, and those who are besieged are Thy priests, I beseech Thee not to hearken to them against these men nor to bring to pass what these men ask Thee to do those others’. And when he had prayed in this manner the villains among the Jews who stood round him stoned him to death (Ant. 14.22–24).

Although Josephus shows considerable interest in weather miracles elsewhere, here Onias is included primarily for his honourable role in a civil war. Onias’s prayer for rain is mentioned only to explain why the Hyrcanus faction thought him capable of praying effectively against their foes (although it is followed by an account of a crop-ruining wind sent by God as a punishment for Onias’s murder and other impieties). If one is justified in identifying Onias with the rabbinic Ḥoni, one may venture two further deductions about him. First, Josephus’s account agrees with that of the Mishnah in associating Ḥoni/Onias with Jerusalem. Secondly, taken together the two accounts suggest that Ḥoni was a public figure: in the Mishnah he is approached to pray for rain; in Josephus he is approached to pray for success in war.

But what kind of figure does this make Ḥoni? He is sometimes referred to in scholarly literature as a ‘charismatic’, but it is seldom made clear why this label should be applied to him. Josephus and the Mishnah agree that Ḥoni merely had his prayers for rain answered on one occasion. This hardly implies that Ḥoni wandered round Palestine regularly performing miracles. Neither is there anything to suggest that Ḥoni possessed ‘charisma’ in the sociological sense of non-institutional authority.

Before reaching our final conclusions on Ḥoni, we should look at the expanded Ḥoni traditions in BT. Ta ‘anith 23a gives an extended account of the Mishnaic story, with additions designed to rabbinize Ḥoni. His standing in a circle is justified through the precedent of Habakkuk. He is provided with disciples to complain that the first two types of rain his prayers evoke are not life-saving. His boastful order to take in the Passover ovens is omitted. When the people and Ḥoni’s disciples ask for him to pray for the rain to stop, Ḥoni’s immediate reply is that he has a tradition that one should not pray on account of an excess of good. He nevertheless commands his disciples to bring him a bullock for a thank-offering and prays a suitably devout prayer, which this time does stop the rain. Ḥoni is thereby transmogrified from the Mishnah’s bumbling magician into a respectable rabbi, even though Simeon b. Shetaḥ is still allowed to complain to him.

Following this heavily revised story there are a series of appreciative remarks, based on scriptural quotations, that the Sanhedrin are said to have sent to Ḥoni. There then follows a story attributed to R. Joḥanan in which Ḥoni falls asleep for 70 years, and wakes to discover that his sons have passed away but that his grandson is still living. No one will believe that he is Ḥoni the Circle-Drawer, even when he presents himself at the Beth ha-midrash where he overhears the scholars saying, ‘The law is as clear to us as in the days of Ḥoni the Circle-Drawer, for whenever he came to the Beth Hamidrash he would settle for the scholars any difficulty that they had.’ Ḥoni is so hurt that no one will believe that he really is Ḥoni that he successfully prays for death. Ḥoni the bumbling sorcerer has now become Ḥoni the accomplished sage.

This leads into a pair of stories about Ḥoni’s grandchildren. The first concerns Abba Ḥilkiah, of whom BT informs us that ‘whenever the world was in need of rain the Rabbis sent a message to him and he prayed and rain fell’. The story then relates how the Rabbis sent a pair of scholars to Ḥilkiah on a particular occasion, and how Ḥilkiah behaved with exemplary piety and modesty. In particular, surmising the purpose of the scholars’ visit without waiting to be told, he repaired to the roof of his house to pray for rain, so that rain might come without his receiving credit for it. The second concerns Ḥanan ha-Neḥba, the son of Ḥoni’s daughter. Whenever there was a drought, the rabbis would send school-children to ask him to pray for rain. In reply he would pray that the Master of the Universe might teach the children to distinguish between the Father who gives rain and the father who does not.

Few of these stories have any historical value for an investigation of Second Temple times. Even if they did, and Ḥoni resembled his grandsons, the picture that would emerge would not be one of a ‘charismatic miracle-worker’ exercising a ministry in any way comparable to that of Jesus, but of a humble and devout man of prayer. Charismatic authority does not run in families; and to use Kahl’s terminology, neither Ḥoni nor his grandsons are portrayed as BNPs or MNPs, but simply as PNPs; their effectiveness derives not from thaumaturgical prowess but from humble piety. Once one sees the Mishnaic account of Ḥoni as addressing third-century concerns rather than describing Second Temple history, there is nothing in the Ḥoni traditions to suggest that Ḥoni belonged to a class of charismatic miracle-workers, or that his single successful prayer for rain in any way indicates the existence of such a class in his time.

Ḥanina ben Dosa

a. Introduction

On the face of it, Ḥanina ben Dosa presents a far closer parallel to Jesus of Nazareth than does Ḥoni the Circle-Drawer. He is commonly taken to have been a rough contemporary of Jesus, and a wide range of miracles are attributed to him. The story in which Ḥanina brings about the healing at a distance of Gamaliel’s son (Ber. 34b) is often cited as resembling that in which Jesus heals the Capernaum nobleman’s son (Jn 4:46–54), not least because in both stories, the supplicants arrive home to discover that the son had been healed just at the moment specified by the healer. It is thus not surprising that Vermes has made considerable use of the figure of Ḥanina to locate Jesus in the category of charismatic Ḥasid.

b. References to Ḥanina in Tannaitic Literature

The Mishnah mentions Ḥanina ben Dosa in three places:

If he that says the Tefillah falls into error it is a bad omen for him; and if he was the agent of the congregation it is a bad omen for them that appointed him, because a man’s agent is like to himself. They tell of R. Ḥanina b. Dosa that he used to pray over the sick and say, ‘This one will live’, or ‘This one will die’. They said to him, ‘How knowest thou?’ He replied, ‘If my prayer is fluent in my mouth I know that he is accepted; and if not I know that he is rejected’ (m. Ber. 5.5).

When R. Ḥanina b. Dosa died the men of [deed] ceased (m. Soṭ 9.15; essentially the same tradition is reproduced at t. Soṭ 15.5).

R. Ḥanina ben Dosa said: He whose fear of sin comes before his wisdom, his wisdom endures; but he whose wisdom comes before his fear of sin, his wisdom does not endure. He used to say: He whose works exceed his wisdom, his wisdom endures; but he whose wisdom exceeds his works, his wisdom does not endure. He used to say: He in whom the spirit of mankind finds pleasure, in him the spirit of God finds pleasure; but he in whom the spirit of mankind finds no pleasure, in him the spirit of God finds no pleasure (m. Ab. 3.10–11).

In addition, the Tosefta narrates the following story about Ḥanina and a lizard:

A.    One who was standing and reciting the Prayer in a camp or in a wide highway

B.    lo, he may move aside to allow an ass, an ass-driver or a wagon-driver to pass in front of him, but he may not interrupt [his recitation of the Prayer].

C.    They related about R. Ḥaninah [sic] b. Dosa that once while he was reciting the Prayer, a poisonous lizard bit him, but he did not interrupt [his recitation].

D.    His students went and found it [the lizard] dead at the entrance to its hole.

E.    They said, ‘Woe to the man who is bitten by a lizard. Woe to the lizard that bit Ben Dosa’ (t. Ber. 3.20).

One might further include the reference to Ḥanina in the Mekhilta of R. Ishmael, which, in commenting on Exod. 18:21, describes Ḥanina as a ‘man of truth’.

The age of a tradition does not necessarily correspond to the age of the writing in which it is first fixed, but it may nevertheless be valuable to enquire what picture of Ḥanina b. Dosa would emerge if we restricted ourselves to Tannaitic literature.

First, there is nothing in this material to indicate when Ḥanina b. Dosa lived. The Mishnah tractate Aboth does not cite its authorities in chronological sequence (this would place Ḥanina after R. Dosethai b. Yannai who lived towards the end of the second century and before R. Dosa b. Harkinas who lived towards the end of the first). Neither are the rabbis in m. Soṭ 9.15 arranged chronologically. The information used to date Ḥanina in the first century, and generally before the fall of the Temple, is taken entirely from Amoraic and Geonic literature.

Secondly, the Tannaitic sources are consistent in calling Ḥanina a rabbi. But thirdly, no halakhoth are attributed to Ḥanina, but merely a handful of wisdom-type sayings putting deeds and fear of sin above wisdom, and suggesting that the way to be right with God is to be right with one’s fellow human being.

Fourthly, there is nothing to indicate that Ḥanina was a miracle-worker, unless this is the meaning of the disputed phrase ‘men of deed’ (see below). The account of Ḥanina’s prayers in m. Ber. 5.5 does not indicate that they were especially efficacious in healing, but only that Ḥanina could tell from the feel of them whether the person prayed for would recover. If there was a miracle in Ḥanina’s survival of snake-bite, there is no indication that Ḥanina performed it; the point of the story is not that Ḥanina is a wonder-worker but that no evil befalls the devout.

Despite the first point, the second and third ones may indicate a date prior to 100 ce for Ḥanina. The sources call him ‘rabbi’, but they do not show him transmitting or engaging with the sayings of any other rabbi. This suggests the title may have been applied to him in retrospect. In the one place where he is addressed in direct speech in the Tannaitic literature, Ḥanina is called simply ‘ben Dosa’ without any formal title.

We are thus left with a shadowy figure, probably of the Second Temple period, who was apparently well-known for his piety, for deeds of an unspecified nature, for his ability to discern how his prayers for the critically ill were being received, and for a few sayings about the priority of deeds over wisdom. Even these meagre traditions suggest that there must have been something to Ḥanina. Otherwise he would not be a person whose remarks would be of interest, or of whom a story about surviving an encounter with a poisonous reptile would be told, or who would have gained a reputation as a notable ‘man of deed’ or ‘man of truth’.

We must next look at what the Amoraic literature (which may well contain Tannaitic traditions) has to say about Ḥanina in order to see if this shadowy outline can be brought into sharper focus.

c. The Talmudic Material about Ḥanina

I have already indicated scepticism about the value of Talmudic material for first-century history. One might nevertheless justify using it on the grounds that the Mishnah does not necessarily represent the earliest stratum of tradition, but may sometimes contain only an echo of traditions preserved more fully as baraitot in the Talmuds. The most obvious objection to this is the length of time over which the oral traditions would have to have been transmitted. Scholars are regularly sceptical about material, not least miracle stories, which appears in the Gospel of Mark roughly four decades after the death of Jesus. In the case of Ḥanina the traditions will have been transmitted over the course of about four centuries.

Doubts about the historical value of this material are scarcely eased when one examines its content. The largest collection of talmudic Ḥanina stories comes in b. Ta’an. 24b–25b. These may be summarized as follows:

a.    Ḥanina is journeying home when it begins to rain. He exclaims to God that he is in distress while the rest of the world is at ease. The rain stops. On his return home he exclaims that he is at ease while the rest of the world is in distress. The rain starts again (b. Ta’an. 24b).

b.    A heavenly voice declares that the whole world draws its existence on account of Ḥanina, while Ḥanina is content to subsist on a single kab of carobs from one Sabbath to another (b. Ta ‘an. 24b).

c.    Ḥanina’s wife throws smoke-making material on her oven to hide the fact that she has nothing to cook for the Sabbath. When a prying neighbour calls to investigate, her oven is miraculously filled with loaves (b. Ta ‘an. 24b–25a).

d.    Ḥanina’s wife complains about their poverty. Ḥanina prays and a hand appears offering him a golden table leg. But he dreams that in the world to come he would be eating at a two-legged table while the pious would be eating at three-legged tables; he therefore prays for the leg to be taken away again (b. Ta ‘an. 25a).

e.    Ḥanina’s daughter is upset because she poured vinegar instead of oil into the Sabbath lamp. Ḥanina tells her that He who commanded the oil to burn will also command the vinegar to burn, and according to a baraita the vinegar burned all day. (b. Ta ‘an. 25a).

f.    Ḥanina’s goats are accused of causing damage. He says that if they are guilty, bears will devour them, but if they are innocent they will each bring a bear home on their horns, which they duly do (b. Ta ‘an. 25a).

g.    A neighbour of Ḥanina named Ikku was building a house whose beams did not reach the walls. At Ḥanina’s orders the beams duly lengthened to protrude one cubit on either side. Polemo subsequently saw the house and was told it was the one R. Ḥanina ben Dosa had covered with beams through his prayer (b. Ta ‘an. 25a).

h.    In answer to the question how Ḥanina kept goats if he was so poor, and why he disobeyed the prohibition against keeping goats in Palestine, an account attributed to R. Phinehas relates that Ḥanina bought the goats with the proceeds of the sales of eggs from hens left near his house. He also sold the hens. When the owner of the hens returned, Ḥanina gave him the goats (b. Ta ‘an. 25a).

To this may be added a number of stories from b. Berakoth and elsewhere:

i.    Villagers troubled by a poisonous reptile ask Ḥanina for help. On being shown the snake’s hole Ḥanina places his heel over it, the reptile bites him and duly dies. Ḥanina carries its carcass back to the school-house and declares ‘See, my sons, it is not the lizard that kills, it is sin that kills!’ whereupon those present declare ‘Woe to the man whom a lizard meets, but woe to the lizard which R. Ḥanina b. Dosa meets!’ (b. Ber. 33a).

j.    R. Gamaliel sends scholars to Ḥanina to ask him to pray for his sick son. Ḥanina prays and announces a cure. On returning to Gamaliel the messengers discover the fever left Gamaliel’s son just when Ḥanina said (b. Ber. 34b).

k.    Joḥanan b. Zakkai asks Ḥanina (his student) to pray for his son. Ḥanina does so and the son lives (b. Ber. 34b).

l.    Rab said, ‘The world was created only for Ahab son of Omri and for R. Ḥanina b. Dosa; for Ahab son of Omri this world, and for R. Ḥanina b. Dosa the future world’ (b. Ber. 61b).

m.    Agrath the demoness meets Ḥanina, but is unable to harm him owing to an announcement made about him in heaven. Ḥanina orders her never to pass through settled regions again, but she pleads for leeway, so Ḥanina allows her to go abroad on Sabbath and Wednesday nights (b. Pes. 112b).

n.    The daughter of Nehuniah the ditch-digger falls into a hole. When Ḥanina is informed he says ‘Peace’ at hourly intervals until after three hours he announces that she has come up. On being asked whether he is a prophet Ḥanina quotes Amos 7:14 and asks whether a righteous man’s offspring should perish as a result of his labour. A saying of R. Aha adds that ‘All the same, his son died of thirst’ (b. B. Qam. 50a).

o.    Ḥanina’s donkey refuses untithed food and returns home (ARNa 8).

This list does not exhaust the talmudic Ḥanina material, but it includes most of it. The Palestinian Talmud gives only three Ḥanina miracle stories: an abbreviated account of Ḥanina and Gamaliel’s son (y. Ber. 7d), another version of Ḥanina and the snake (y. Ber. 7a), and a story about Ḥanina’s dinner table collapsing to prevent his eating untithed food (y. Dem. 22a). It is curious that the bulk of the Ḥanina material appears in the Babylonian collection, whereas the Talmud compiled in Palestine, where Ḥanina actually lived and where one might have expected the traditions about him to be circulating, is remarkably restrained. Although it falls far short of proof, this suggests that the Babylonian Amoraim were interested in collecting and developing Ḥanina legends for reasons that had little to do with concerns for first-century history.

This suspicion is hardly allayed by the content of the tradition. Items (c)—(g) and (o) are hardly accounts of historical events; (a) is similarly dubious, whereas (m) looks plainly legendary; (b), (1) and other similar sayings about Ḥanina reflect rabbinic opinions of Ḥanina rather than historical information about him; (i) is a further development of the tradition about Ḥanina and the snake, rather than an independent account of a separate incident, and thus illustrates the fluidity of the Ḥanina tradition. This only leaves items (h), (j), (k) and (n) as plausibly transmitting some kind of reliable historical data, and even then the historicity of this material is far from assured.

Even if this material could be taken at face value, the picture that would emerge is not particularly similar to Jesus. The Jesus of the Gospels is presented as engaged in a wandering ministry of preaching, healing, and exorcism, announcing the kingdom of God and occasionally performing spectacular signs that point to the significance of his person. In BT Ḥanina is portrayed as a poor, pious rabbi, living at home with his wife and daughter, occasionally called upon to help in cases of accident, fever and venomous reptiles, and occasionally assisted by bizarre divine interventions. It is only the two healing stories and the encounter with Agrath that suggest any similarity with Jesus, which is why we must examine these particular accounts in more detail. But it is first necessary to question the general historical value of this material a little further.

This material is not primarily concerned with historiography. It reflects other interests. Baruch Bokser points out that the rabbinic literature was used for didactic purposes, which may well result in a story’s being used for a purpose quite different from its original one. Bokser goes on to demonstrate this in some detail through the varying account of Ḥanina and the snake. We have seen a similar process at work with Ḥoni the Circle-Drawer.

There appears to have been a move in rabbinic thought from suspicion of miracles to cautious acceptance of miracle-workers provided their powers were based on knowledge of the law and its pious observance. Some writers see this change in terms of the status and authority of the rabbi. The rabbis of the Tannaitic period were anxious to establish their authority and were suspicious of potential rivals such as charismatic prophets and miracle-workers. By Amoraic times, however, the social standing of rabbis (at least among Babylonian Jews) had become such that, not only was their status secure, but miracles might be positively expected from them. Alternatively, these developments might be seen in terms of the type of piety the rabbis promoted at different times. Only the gifted few could be miracle-workers. Miracle working was thus an unacceptable model of piety for the Tannaim, who wanted something that everyone could aspire to, but became more acceptable in the changed circumstances of their successors.30

Either explanation may be correct. They both account for the change of attitude from the Tannaitic literature to BT in terms of the changing situation of the rabbis. They both therefore suggest that the Ḥanina material in rabbinic literature tells us more about the circumstances under which the literature was redacted than about any historical first-century figure.

This does not mean that Ḥanina should be dismissed altogether. The reference to Polemo (a contemporary of Judah-ha Nasi) at b. Ta ‘an. 25a suggests that a Galilean tradition of Ḥanina’s miracle working had arisen by the end of the second century. Pace Vermes, this does not show that the Ḥanina traditions were ‘fully developed’ by this stage (since we have seen good reasons to suppose they continued to develop into the Amoraic period). But it is hardly likely that the reputation reflected in the Ikku story and the other Talmudic material grew up solely on the basis of the shadowy figure glimpsed through the Mishnah and Tosefta. The material in BT thus strengthens the case for supposing that Ḥanina was a notable figure in his own time. But what was he notable for? Sean Freyne effectively proposes that he gained a reputation as a PNP based on ‘intimate knowledge of the divine action rather than miraculous powers which he himself possessed’.33 This suggestion seems as good as any, but still leaves us with no clear picture what sort of figure the historical Ḥanina was. This is a question we shall return to when we discuss what was meant by calling him a ‘man of deed’, but we must first look more closely at the healings brought about through his prayers.

d. Ḥanina’s Healings

The stories in which Ḥanina b. Dosa prays for the sons of R. Gamaliel and R. Joḥanan provide the closest parallels to Jesus’ reported miracles. These two stories appear consecutively in Ber. 34b:

Our Rabbis taught: Once the son of R. Gamaliel fell ill. He sent two scholars to R. Ḥanina b. Dosa to ask him to pray for him. When he saw them he went up to an upper chamber and prayed for him. When he came down he said to them: Go, the fever has left him: They said to him: Are you a prophet? He replied: I am neither a prophet nor the son of a prophet, but I learnt this from experience. If my prayer is fluent in my mouth, I know that he is accepted: but if not, I know that he is rejected. They sat down and made a note of the exact moment. When they came to R. Gamaliel, he said to them: By the temple service! You have not been a moment too soon or too late, but so it happened: at that very moment the fever left him and he asked for water to drink.

On another occasion it happened that R. Ḥanina b. Dosa went to study Torah with R. Joḥanan ben Zakkai. The son of R. Joḥanan ben Zakkai fell ill. He said to him: Ḥanina my son, pray for him that he may live. He put his head between his knees and prayed for him and he lived. Said R. Joḥanan ben Zakkai: if Ben Zakkai had stuck his head between his knees for the whole day, no notice would have been taken of him. Said his wife to him: Is Ḥanina greater than you are? He replied to her: No; but he is like a servant before the king, and I am like a nobleman before a king.

These stories place Ḥanina in close association with two great proto-rabbinic figures of the first century (whichever Gamaliel is meant). This suggests a deliberate attempt to portray Ḥanina as operating within the rabbinic fold. The second story confirms this impression. Ḥanina is made into a rabbi by being presented as a pupil of the great teacher Joḥanan b. Zakkai. This is suspect as history, since the historical Ḥanina was almost certainly not a rabbi. But this story does not merely associate Ḥanina with Joḥanan, it subordinates the miracle-worker to the rabbi. Ḥanina is expressly Joḥanan’s pupil, and Joḥanan is given the last word. The ‘miracle-worker’ is effective in interceding for Joḥanan’s son, but this only makes him like a favoured slave who has access to the king whenever he likes, whereas Joḥanan is like a nobleman who has to appear before the king at set times in accordance with court protocol. The message is clear: Ḥanina’s prowess at praying does not make him superior to the great rabbi. The story thus legitimates the miracle-worker, but only if he operates on rabbinic terms. It probably has no historical basis at all.

The story about Ḥanina and Joḥanan may well have been devised to qualify the story about Ḥanina and Gamaliel. If so, this may indicate that the Amoraim did not tamper unduly with the Gamaliel story (since they were able to achieve their objective by appending the Joḥanan account). Moreover, the baraita about Ḥanina and Gamaliel contains the saying of Ḥanina given at m. Ber. 5.5, which indicates that the entire pericope may go back to Tannaitic times. It can be argued that Ḥanina’s saying in the Mishnah presupposes the Talmudic tale. The alternative is that the story contained in b. Ber. 34b grew from the saying in m. Ber. 5.5. But since some story must lie behind the saying attributed to Ḥanina in the Mishnah, there is no reason why it should not have resembled that narrated in BT. This falls short of demonstrating that BT’s story goes back to the historical Ḥanina, but it suggests that this story has the highest claim of any Ḥanina story to do so.

Since this story resembles that of Jesus’ healing the Capernaum nobleman’s son, it will now be expedient to compare the two. The similarities between these stories are: (a) both tell of a request to the healer from a person of note which (b) concerns that person’s son who (c) is suffering from fever; moreover (d) the son’s recovery is announced by the healer from a distance and (e) when the petitioners return home they find that the son recovered at precisely the time the healer announced and (f) that recovery is described as the fever leaving him.

These similarities are sufficiently impressive to lend support to a first-century origin for the BT story. But there are also significant differences. The main one is that whereas in the Johannine story Jesus is a BNP, in the rabbinic story Ḥanina is merely a PNP.40 Whereas Jesus heals at a distance, Ḥanina merely prays at a distance; in the Ḥanina story it is God who actually heals, presumably at no lesser or greater distance than God usually does. That a human agent should heal from a distance is remarkable; but praying for someone’s healing from a distance is surely commonplace. What is remarkable about Ḥanina is that he can tell straight away whether his intercession has been accepted. But this points to another difference between Jesus and Ḥanina. Mark 6:5 notwithstanding, nowhere in the Gospels is Jesus’ ability to heal seriously thrown into question; Ḥanina, on the other hand, apparently has no idea whether his prayer will prove efficacious until he prays it. This story does not therefore portray Ḥanina as a miraculously gifted healer, but rather as someone who is especially proficient at prayer. Conversely, the Johannine pericope is greatly occupied with the question of faith, a theme that scarcely surfaces in the Ḥanina story.

A more general difference between Ḥanina’s miracles and Jesus’ miracles as presented in the Synoptic tradition is that there is no indication that Ḥanina is engaged in a struggle aimed at defeating the powers of darkness. There is, for example, no indication that Ḥanina performed so much as a single exorcism, whereas the Synoptic Gospels present casting out demons as a major element in Jesus’ ministry. The single instance in which Ḥanina is said to have confronted a demon provides no exception, as we shall now go on to see.

e. Ḥanina and the Princess of Demons

The account of Ḥanina and Agrath occurs at b. Pes. 112b, in the context of a series of prohibitions, some prudential, others more superstitious. The Ḥanina pericope explains why the prohibition against going out alone at nights applies only to Wednesdays and Saturdays. It is presumably to explain this rather than to provide historical information that the story is told. It also appears anxious to stress Heaven’s endorsement of Ḥanina’s learning, which suggests an advanced stage in the rabbinization of Ḥanina.

As in the case of Jesus’ encounter with the ‘Legion’ demon, the demoness recognizes Ḥanina’s authority but manages to secure a concession from him. But here the resemblance ends. The demons Jesus encounters are always within some unfortunate human being, and Jesus’ response to them is always to cast them out. There is no suggestion that Agrath was in possession of a human victim’s body when she encountered Ḥanina; the story rather suggests that he came upon her as a free-standing demoness in human form. The closest parallels are not the exorcism stories of the Gospels, but the rabbinic stories that Vermes adduces as parallels.

R. Meir used to scoff at transgressors. One day Satan appeared to him in the guise of a woman on the opposite bank of the river. As there was no ferry, he seized the rope and proceeded across. When he had reached half way along the rope, he [Satan] let him go saying, ‘Had they not proclaimed in Heaven, “Take heed of R. Meir and his learning”, I would have valued your life at two ma ‘ahs‘.

R. Akiba used to scoff at transgressors. One day Satan appeared to him as a woman on the top of a palm tree. Grasping the tree, he went climbing up: but when he reached half-way up the tree he [Satan] let him go, saying, ‘Had they not proclaimed in Heaven, “Take heed of R. Akiba and his learning,” I would have valued your life at two ma ‘ahs’ (b. Qid. 81a).

The point of these stories seems to be that since even such great legal authorities as Meir and Akiba can succumb to (sexual) temptation, the ordinary man had better not complacently assume himself immune. If these stories provide the correct background for understanding that about Agrath, then two deductions may be made: first, that Agrath appears to Ḥanina in her own (presumably alluring) female form, and second, that Ḥanina was transgressing rabbinic propriety and subjecting himself to temptation by walking out alone at night.

There is no eschatological dimension to this encounter. Like Jesus, Ḥanina can command obedience from the demon, but neither Ḥanina nor Agrath shows any signs of supposing that this is the opening move in the final destruction of all demonic forces. Agrath is simply obliged to obey Heaven’s ruling concerning Ḥanina, a ruling that places him on a par with other great rabbis.

This story is not an exorcism. Ḥanina does not expel Agrath from a possessed human person. He does not even expel her from a particular place (but instead restricts her access to inhabited places in general). Both the legendary nature of this account and its function within the rabbinic corpus indicate that it cannot be used to show that there was ever any tradition of Ḥanina performing exorcisms.

f. ‘Men of Deed’ and Ḥanina’s Type

The Mishnah states that when R. Ḥanina ben Dosa died, the men of deed ceased (m. Soṭ. 9.15). This is one of a group of sayings of the form ‘When Rabbi X died, Virtue A ceased.’ These cannot be intended to mean that with the death of Rabbi X virtue A literally vanished from the face of the Earth, but rather that Rabbi X was the outstanding exemplar of virtue A. According to the mediaeval rabbinic commentator Rashi, and to many modern scholars (such as Vermes and Freyne), ‘man of deed’ (איש מעשה) means miracle-worker. If this is correct then, despite all that I have said up to now, not only might Ḥanina have been a notable miracle-worker, but he would be but one example of a (possibly sizeable) class of such miracle-workers.

Yet there are good reasons for supposing that that ‘man of deed’ (איש מעשה) in m. Soṭ. 9.15 does not mean ‘miracle-worker’. Adolf Büchler argues that, by itself, מעשה never means ‘miracle’ unless qualified by נסים or a synonym thereof. In his view, Rashi arrived at his opinion on the basis of the application of ‘man of deed’ to Ḥanina, of whom many miracle-stories are narrated. But it may be deeds of a quite different sort that earned Ḥanina this title. Büchler goes on to cite a number of passages where מעשה means no more than ‘deed’, together with others where it refers to the practice of the positive precepts of the Torah, or to good deeds. It is in one of these latter senses that he supposes the term מעשה to have been applied to Ḥanina. S. Safrai also takes ‘deed’ to mean ‘good deed’ rather than ‘miracle’. In his view, the אנשי מעשה were Ḥasidim who were ‘active in human society’ and especially concerned with such public good works as ‘redemption of captives, the digging of cisterns for the benefit of wayfarers, the restoration of lost property, the consolation of mourners, the giving of alms etc.’

Vermes concedes that by itself, ‘man of deed’ does not necessarily mean ‘miracle-worker’, but nevertheless proposes a number of counterarguments in the case of Ḥanina ben Dosa. One is that the closest parallels to Ḥanina’s (miraculous) deeds are found not in the Talmud, but the New Testament. But this is precisely the point at issue. Another is that the New Testament regularly refers to Jesus’ miracles as ‘works’ (ἔργα). But one can hardly appeal to the Greek usage of the New Testament to determine the Hebrew usage of the Mishnah and Tosefta, and in any case the New Testament does not refer to Jesus as an ἀνθρώπος ἔγρων or ἀνήρ ἔργων. The description of Jesus’ miracles as ‘works’ is characteristically Johannine, even assuming that ἔργα in John refers exclusively to Jesus’ miracles. Although Lk. 24:19 describes Jesus as ‘a prophet mighty in deed and word’, ‘deed’ is there qualified by close association with ‘prophet’ and ‘mighty’.

A third argument is that Josephus describes Jesus as a doer of ‘marvellous deeds’ (παραδόξων ἔργων, Ant. 18.63), but here Josephus is obliged to qualify ἔργων with παραδόξων, one of his characteristic words for the miraculous. This leaves only one argument worth considering, namely that Ḥanina’s deeds are described not as charitable acts but as miracles; there is no evidence in the sources to suggest that Ḥanina was primarily ‘a social worker or promoter of public welfare’.

Vermes does have a point here. The incidents Büchler cites to show that Ḥanina was a major doer of good works are far from convincing, not least because they are of no more historical value than the miracle stories (of which they frequently form a part). On the other hand, one should be wary of taking the outstanding feature of a single individual as the defining characteristic of the class to which he is said to belong.51 In other words, the fact that the most notable needs associated with Ḥanina may have been miraculous does not show that this need have been the case with all other ‘men of deed’.

The obvious inference to be drawn is that the phrase ‘men of deed’ means no more than it says, men who were most notable for their (presumably good or remarkable) actions rather than, say, for their piety or learning. The phrase ‘men of deed’ might then include ‘miracle-workers’, but by no means be restricted to them. Thus the application of ‘man of deed’ to Ḥanina can be used neither to determine that he was a miracle-worker (the argument would in any case be circular) nor to show that he belonged to a class of miracle-workers. It suggests merely that Ḥanina was one of the more notable of a number of people renowned for notable deeds.

There is little reason to suppose that either Ḥanina in particular or miracle-workers in general also belonged to the Ḥasidim. Even if later Jewish literature did tend to conflate the two, this may well have been a parallel to the tendency in Graeco-Roman religion to identify roles that had formerly been separate (as in the late convergence of the philosopher and the miracle-worker into the θεῖος ἀνήρ). Only at t. Ta ‘an. 2.13 is there a story of a Ḥasid who is asked to bring rain and then stop it, and in that instance the title probably refers to his righteousness rather than his rain-making. There is, after all, no reason why an individual Ḥasid should not function in several different roles; but then a role peculiar to a particular Ḥasid may not be taken as defining the type. Even if Ḥanina was a miracle-working Ḥasid (which there is little reason to suppose), this would be no reason to postulate the existence of a whole class of miracle-working Ḥasidim.

Neither can Ḥanina be used to establish the existence of any other type, such as that of Galilean charismatic miracle-worker. To establish the existence of such a type, one would need to identify several persons sharing relevantly similar characteristics and performing relevantly similar social roles. In this instance, one must leave Jesus aside, since he cannot without circularity be used to establish the existence of a type to which he is subsequently to be assimilated. This leaves us with Ḥanina, Ḥoni, Ḥoni’s grandchildren and perhaps a few later rabbis. But there is no evidence that these people were relevantly similar. Ḥoni once prayed for rain and subsequently got caught up in a civil war. Ḥanina seems to have been a devout man of prayer and may perhaps have functioned as a folk-healer (for which see Chapter 13). It is unclear whether we have any reliable information about Ḥoni’s grandchildren at all. These are not sound foundations on which to construct a class of charismatic miracle-workers. They are not sound foundations on which to build any class.

In any case, if Jesus is the paradigm of a charismatic miracle-worker then there is little to suggest that Ḥanina fits the same description. Unlike Jesus, Ḥanina does not seem to have been an exorcist, he did not obviously fall foul of the establishment, there is nothing to suggest that he performed a wandering ministry of healing (let alone teaching), and, so far as one can tell, he was a ‘miracle-worker’ only in the sense of being a PNP, and not an MNP, let alone a BNP. The historical Ḥanina may have been notable in his time, perhaps greatly loved and admired in his community, but he has done little to help us understand the miracles of Jesus.


The conclusions of this chapter have been mainly negative, namely that Ḥanina ben Dosa and Ḥoni the Circle Drawer may not provide such significant parallels to Jesus of Nazareth as has sometimes been suggested. But to end on a more positive note, there is one important strand of Vermes’s treatment of this topic that remains entirely valid. As he points out (and as was argued in the previous chapters), Judaism often associated miracle-working and prophets, not least the prophets Elijah and Elisha. Moreover, Vermes makes a good case to the effect that rabbinic traditions associated both Ḥoni and Ḥanina with Elijah, and that the Gospel traditions sometimes make the same association between Jesus and the Elijah/Elisha cycles. The traditions about Ḥoni and Ḥanina therefore support the thesis that Judaism tended to associate miracle-working with prophets such as Elijah and Elisha, and even though the rabbinic traditions about Ḥoni and Ḥanina may be later (at least in the form that we now have them) than the traditions about Jesus, the other evidence presented in this study indicates that this association with miracle-working and the biblical prophets persisted over a long time. To that extent, the traditions about Ḥoni and Ḥanina are relevant to ascertaining how Jesus’ miracles may have been understood; but note that this remains distinct from saying that the historical Jesus belonged to the same general type as the historical Ḥoni or the historical Ḥanina.

A class of would-be miracle-workers close to the time of Jesus about whom we appear to have more reliable historical information are the ‘sign prophets’ described by Josephus. These will form the subject of the next chapter.

Chapter 11

The Sign Prophets


Josephus’s description of the so-called ‘sign prophets’ indicates that they were active in first-century ce Palestine and that promising at least one miracle was a significant part of their activity. Moreover, Jesus shared at least some superficial characteristics with these figures: he attracted a crowd of followers; he promised a decisive divine intervention; he claimed to be God’s spokesman; he was executed by the Roman authorities. Again, Luke appears to regard classification of Jesus with the sign prophets as superficially plausible. At Acts 5:36 Gamaliel likens Jesus to Theudas, and at Acts 21:38 Paul is taken by a Roman officer to be ‘the Egyptian’, presumably the same character as ‘the Egyptian’ described by Josephus at War 2.261–63 and Ant. 20.169. In John’s account of the feeding of the five thousand the crowd identifies Jesus as the Prophet and even tries to make him king. Finally, Mk 13:22 warns that ‘False Christs and false prophets will arise and show signs and wonders, to lead astray, if possible, the elect.’ Matthew 24:26 further indicates that these people might appear ‘in the wilderness’. These figures sound not unlike Josephus’s sign prophets, regarded as ‘false prophets’ by the Evangelists and Josephus alike. But if there is some danger that they may succeed in leading astray the elect, they must bear some resemblance to the genuine article. It thus seems worthwhile to examine these ‘sign prophets’ more closely, as part of the Jewish context of Jesus’ miracles.

The Accounts of the Sign Prophets

a. Theudas

Here it will be helpful to set out both Josephus’s account in Ant. 20.97–99 and the briefer account supplied by Gamaliel in Acts 5:36. According to Josephus,

During the period when Fadus was procurator of Judaea, a certain impostor (γοής) named Theudas persuaded the majority of the masses to take up their possessions and to follow him to the Jordan River. He stated that he was a prophet and that at his command the river would be parted and would provide them an easy passage. With this talk he deceived many. Fadus, however, did not permit them to reap the fruit of their folly, but sent against them a squadron of cavalry. These fell upon them unexpectedly, slew many of them and took many prisoners. Theudas himself was captured, whereupon they cut off his head and brought it to Jerusalem. These, then, are the events that befell the Jews during the time that Cuspius Fadus was procurator.

According to Acts 5:36,

For before these days Theudas arose, giving himself out to be somebody (λέγων εἶναι τινα ἑαντόν), and a number of men, about four hundred, joined him; but he was slain and all who followed him were dispersed and came to nothing.

One wonders whether Luke and Josephus can be talking about the same person, since Gamaliel’s speech places Theudas prior to Judas the Galilean, i.e. before 6 ce, whereas Josephus places him during the governorship of Fadus, around 45 or 46 ce, at least ten years later than the speech in which Gamaliel is made to refer to him. Josephus sometimes narrates events out of sequence, but he is unlikely to have placed Theudas under the wrong governor. On the other hand, Luke’s chronology of events at the turn of the era is known to be a little confused. Moreover, Theudas is not such a common name as to make it probable that there were two notable Jewish rebels thus called. On balance, it seems most likely that Luke and Josephus are talking about the same person.

Parting the Jordan most obviously recalls the exodus and conquest traditions. Luke and Josephus both state that Theudas attracted a substantial body of followers. Luke’s ‘about four hundred’ looks more plausible than Josephus’s ‘majority of the masses’ (even if it simply reflects Luke’s fondness for numbering the size of crowds). A squadron of cavalry might suffice to deal with four hundred dupes, but the ‘majority of the masses’ would probably have required the odd legion or two, and would also imply that the Jewish revolt had broken out 20 years before its time. To attract such a following one might claim to be the Davidic messiah, or set oneself up as a king, but there is nothing in either Josephus’s or Luke’s account to suggest that Theudas did either of these things (as opposed to other figures reported by Josephus who were royal pretenders). It is not even clear if Theudas’s followers were armed.4 ‘Prophet’ thus seems an entirely plausible category for Theudas.

It is interesting that, according to Josephus, Theudas persuaded his supporters ‘to take up their possessions’ and follow him. This looks less like a military operation than an attempt to recreate the exodus. It is not clear, however, in which direction Theudas proposed to cross the Jordan. If from west to east, then he might have been leading an exodus from Roman rule; if from west to east he might have been hoping to re-enact the conquest of the land. The latter would have been more obviously threatening but either may have provoked a Roman response.

It is also unclear what exactly Josephus intends by λόης. The word can mean ‘impostor’ as the Loeb edition translates it, but also ‘sorcerer’, ‘juggler’, ‘swindler’ or ‘cheat’. The English word ‘charlatan’ might best capture the range of meanings, but that does not tell us whether Josephus uses the word principally because he thought Theudas was a magician, or principally because he thought he was a deceiver.

Neither is it clear how Theudas expected to deliver on his promise to part the Jordan. Josephus suggests that he failed, but then Josephus is a hostile witness. Even so, it is hardly likely that Theudas managed to part the Jordan by magical or miraculous means. If he was simply a charlatan, he will have needed to arrange some other means of bringing about the desired effect. Since it is hard to imagine how, the most likely conclusion is that Theudas genuinely believed himself to be a prophet called by God, and supposed that God would part the waters as he once had for Moses, Joshua, Elijah and Elisha.7

b. The Egyptian

Josephus gives two accounts of an Egyptian deceiver, one in the Jewish Antiquities and the other in the Jewish War. Since these differ in several important details, it will be as well to set them out side by side:

War 2.261–63

Ant. 20.169–72

A still worse blow was dealt at the Jews by the Egyptian false prophet (ψευδοπροφήτης). A charlatan (ἄνθρωπος γόης), who had gained for himself the reputation of a prophet, this man appeared in the country, collected a following of about thirty thousand dupes, and led them by a circuitous route from the desert to the mount called the mount of Olives. From there he proposed to force an entrance into Jerusalem and, overpowering the Roman garrison, to set himself up as tyrant of the people, employing those who poured in with him as his bodyguard. His attack was anticipated by Felix, who went to meet him with the Roman heavy infantry, the whole population joining him in the defence. The outcome of the ensuing engagement was that the Egyptian escaped with a few of his followers; most of his force were killed or taken prisoners; the remainder dispersed and stealthily escaped to their several homes.

At this time there came to Jerusalem from Egypt a man who declared that he was a prophet and advised the masses of the common people to go out with him to the mountain called the Mount of Olives, which lies opposite the city at a distance of five furlongs. For he asserted that he wished to demonstrate there that at his command Jerusalem’s walls would fall down, through which he promised to provide them an entrance into the city. When Felix heard of this he ordered his soldiers to take up their arms. Setting out from Jerusalem with a large force of cavalry and infantry, he fell upon the Egyptian and his followers, slaying four hundred of them and taking two hundred prisoners. The Egyptian himself escaped from the battle and disappeared.

To this should be added the brief reference in Acts 21:38, where the tribune asks Paul, ‘Are you not the Egyptian, then, who recently stirred up a revolt and led the four thousand men of the Assassins (σικαρίων) out into the wilderness?’

All three accounts agree in describing a figure from Egypt who led some kind of rebellious activity. There is also agreement that this man had escaped Roman retribution: Josephus says so on both occasions, and Luke’s tribune clearly supposes that the Egyptian is still at large. Thereafter the situation becomes more confused.

Josephus’s two accounts agree that the Egyptian claimed to be a prophet and that he led his followers by way of the Mount of Olives with the intention of capturing Jerusalem. Both agree that this plan was forestalled by Felix’s use of Roman arms. According to War Felix employed Roman heavy infantry, supported by the entire population of Jerusalem. In the Antiquities Felix deploys a mixed force of infantry and cavalry, without any explicit assistance from the citizens of Jerusalem. According to War the Egyptian was followed by about ‘thirty thousand dupes’, most of whom were captured or killed. Antiquities does not state the total figure but indicates that those captured or killed totalled about six hundred, which, if taken in conjunction with War’s statement that only a minority of the Egyptian’s followers escaped, would suggest a total force of no more than a thousand. Luke’s figure of four thousand sicarii is somewhere between the two.

The fact that Luke mentions sicarii is odd. Josephus first introduces the sicarii at War 2.254 (not long before he discusses the Egyptian at War 2.261). Although this makes the presence of sicarii chronologically plausible, it is strange that Josephus does not describe the Egyptian’s followers as sicarii if that was what he considered them to be. In the intervening passage (War 2.258–60) Josephus is quite happy to talk about deceivers and impostors leading men out into the desert with revolutionary intentions (there to be quelled by Felix and Roman arms), and to describe these impostors as being just as pernicious as the sicarii. There is thus no reason why Josephus should have wished to disguise the presence of sicarii in the force with which the Egyptian proposed to storm Jerusalem (although this would be contrary to the normal modus operandi of the sicarii as Josephus describes them).

A further oddity is that Luke’s tribune speaks of the Egyptian leading his sicarii into the desert (with no mention of Jerusalem) whereas in War the Egyptian leads his followers from the desert. It looks as if what Luke has the tribune say in Acts 21:38 is a muddled conflation of War 2.254–63, in which Josephus first talks about the sicarii, then about impostors who lead people out into the desert, and then about the Egyptian.

If one leaves aside the complication of the account in Acts, one still has the differences between Antiquities and War to account for. Gray suggests that the account in Antiquities is more likely to be accurate on the grounds that Josephus tends to militarize the sign prophets in War, in order to associate them with the rebels and others whom he wishes to marginalize. This seems to be broadly correct. For example, at War 2.254–63 Josephus moves rapidly from the sicarii through a somewhat amorphous band of ‘deceivers and impostors’ who incite the multitude ‘under the pretence of divine inspiration’ to his account of the Egyptian. This is followed by a description of how ‘The impostors (γόητες) and brigands, banding together, incited numbers to revolt’ (War 2.264). There is a parallel sequence in Ant. 20.160–72, but the details are a little different. First, Felix bribes Doras, a native of Jerusalem, to bring in brigands (probably sicarii to judge by their modus operandi) to assassinate Jonathan the high priest (160–66). Then there is an account of impostors and deceivers calling on the mob to follow them into the desert, where they will be shown signs and wonders (167–68). It is in this context that the Egyptian is described (169–71). That he claims to be a prophet who will cause the walls of Jerusalem to collapse suggests that he is the most notable example of the impostors and deceivers Josephus has just referred to. Finally, Josephus appends a brief account of brigands who incite the populace to rebel, and menace those who are not inclined to be incited willingly (172). As Gray observes, Antiquities lacks the tendency to associate and conflate the brigands and charlatans that is found in War. Since they appear to be two rather different types of people, this separation of roles in Antiquities is prima facie more plausible than the conflation of roles in War.

Other details of the two accounts also tend to make that of Antiquities look more plausible than that of War. For one thing, Antiquities suggests that the Egyptian’s following was on a relatively modest scale. Had he in fact turned up outside Jerusalem with thirty thousand followers willing to do battle with the Romans as War states, we should have to conclude that the Jewish revolt started under Felix. The notice in War that the whole population of Jerusalem assisted Felix in its defence is probably an apologetic device designed to distance the sane majority of Jews from the revolutionary fervour of the lunatic fringe. Again, even in War the Egyptian is described as a false prophet and a charlatan (rather than a ‘brigand’ or self-made ‘king’); it is not really consistent with this that he should be portrayed as the leader of a band of would-be warriors prepared to embark on a siege. The terms ‘false prophet’ and ‘charlatan’ are more plausibly applied to someone who behaves like the Egyptian of Antiquities.

If one accepts the account of Ant. 20.169–72 then the Egyptian was almost certainly hoping for a miracle akin to the walls of Jericho collapsing for Joshua. He presumably supposed that once the walls of Jerusalem came tumbling down, its demoralized garrison would be overcome without much of a struggle. It is unclear if the Egyptian’s followers were even armed. That Felix should have suppressed them by force is nevertheless hardly surprising.

The one thing that is slightly surprising is that Luke’s tribune should jump to the conclusion that Paul was this Egyptian, since there seems to have been nothing in the Temple riot in which Paul found himself embroiled to suggest any obvious link. It may be that Luke was simply following through a parallelism between Jesus and Paul here: just as Gamaliel likened Jesus to Theudas so the tribune likened Paul to the other most outstanding of the prophetic deceivers, the Egyptian. If Luke wanted to make a point of this comparison, it might explain why he took liberties with the dating of Theudas. His point might be that whereas moderately sympathetic outsiders such as Gamaliel and Claudius Lysias could mistake the Christian movement for the general run of prophetic enthusiasts that plagued Palestine at the time, they were plainly mistaken. This point nevertheless concedes that the mistake was possible to make.

c. The Charlatans’ Collective under Felix

Brief mention has already been made of the group of ‘impostors’ and ‘deceivers’ who appear under Felix, and whom Josephus introduces just before his account of the Egyptian. They appear in both Antiquities and War and it will again be helpful to set out the parallel accounts in parallel columns:

War 2.258–60

Ant. 20.167–68

Besides these there arose another body of villains, with purer hands but more impious intentions, who no less than the assassins ruined the peace of the city. Deceivers and impostors (πλάνοι γὰρ ἄνθρωποι καὶ ἀπατεῶνες), under the pretence of divine inspiration fostering revolutionary changes, they persuaded the multitude to act like madmen (δαιμονᾶν), and led them out into the desert under the belief that God would there give them tokens of deliverance (σημεῖα ἐλευθερίας). Against them Felix, regarding this as but the preliminary to insurrection, sent a body of cavalry and heavy-armed infantry, and put a large number to the sword.

With such pollution did the deeds of the brigands infect the city. Moreover, impostors and deceivers (οἱ δὲ γόητες καὶ ἀπατεῶνες) called upon the mob to follow them into the desert. For they said that they would show them unmistakable marvels and signs (τέρατα καὶ σημεῖα) that would be wrought in harmony with God’s design. Many were, in fact, persuaded and paid the penalty of their folly; for they were brought before Felix and he punished them.

These two accounts substantially agree, except that Antiquities is vaguer about the fate of those who followed the ‘deceivers and impostors’. Antiquities fails to mention that Felix sent a military expedition to deal with them, but perhaps this is implied by their being brought before him. The account in Antiquities could, perhaps, be taken to refer to a series of impostors leading a series of followers out into the desert, but that in War clearly suggests a single group of impostors leading a single body of persons and dealt with on a single occasion.

Gray suggests that Josephus has again militarized the account in War, creating the impression that an armed revolt was being suppressed, an impression lacking in Antiquities. This is only partly correct. The account in War is more ‘military’ to the extent that it mentions a heavily armed Roman force, but even on Antiquities’s account some kind of force must have been employed to arrest ‘many’ of the mob and bring them before Felix. Gray may be right in supposing that Josephus tries to make the readers of War imagine the mob to have been armed, but his text does not actually say so. In both War and Antiquities the focus of the mob’s expectation is on the miracles to be performed in the desert, not on any proposed military action. Even on the War account Felix’s response does not necessarily imply that the mob comprised armed insurrectionists; as Morton Smith wryly remarks, ‘All these gatherings the Romans put down by military force, but this tells us nothing of the teachings involved; it was merely the Roman method of discouraging illegal assembly.’ Thus while Gray has correctly identified a difference in tendency between the Antiquities and War accounts, it is not so sharply drawn as she suggests. It is rather that War allows the reader to suppose that the Romans were putting down an armed insurrection more than Antiquities does. Perhaps War also exaggerates the Roman troop deployment. In essence, though, Antiquities and War tell the same story.

Several features of Josephus’s terminology are noteworthy here. Since the phrases πλάνοι γὰρ ἄνθρωποι καὶ ἀπατεῶνες (War) and οἱ δὲ γόητες καὶ ἀπατεῶνες (Ant.) are used to refer to the same group of people, Josephus is apparently using γόητες as a synonym of πλόνοι. It thus appears to be the meaning ‘impostor’, ‘deceiver’ or ‘charlatan’ that is uppermost in his mind rather than ‘magician’. On the other hand, Josephus does not refer to this particular group of charlatans as ‘prophets’ or ‘false prophets’. He does say that they acted ‘under the pretence of divine inspiration’, but that could equally apply to a would-be charismatic prophet or a would-be charismatic king. The word ‘pretence’ doubtless represents Josephus’s own value-judgment; the charismatics in question may well have believed their inspiration to be genuine. That they act together without any one of them emerging as the obvious leader is a curious feature of this account.

Antiquities and War concur in having these charismatic impostors lead the mob out into the desert. Gray correctly warns against jumping to conclusions about the significance of this: ‘desert’ or ‘wilderness’ does not necessarily imply a reference to the exodus traditions; ‘there were apparently many different reasons for making an exodus into the real wilderness: it was the home of bandits, some members of the armed resistance, and quietistic religious ascetics, among others’. Nevertheless certain features of the narrative make a reference to the exodus-conquest tradition a prima facie probability here, in particular the promise of ‘signs of liberation’ or ‘unmistakable wonders and signs’. The phrase ‘wonders and signs’ is admittedly a cliché, but it is used in the Septuagint especially of the miracles accompanying the exodus events. The association is surely strengthened when such miracles are expected ‘in the wilderness’. Moreover, traditions concerned with Israel’s liberation from foreign oppression and consequent possession of her land would be highly pertinent to a situation of revolutionary fervour in Roman-occupied Palestine. War describes these promised miracles as ‘signs of liberation’ (σημεῖα ἐλευθερίας), a phrase similar to that Josephus uses at Ant. 2.327 when the Hebrews, trapped between the Red Sea and the pursuing Egyptians, are forgetful of the signs already wrought by God for their liberation (πρὸς τὴν ἐλευθερίαν αὑτοῖς σημείων). It has already been argued (in Chapter 2) that this most naturally refers back to the plagues that secured the Israelites’ release, despite the fact that throughout the first half of Antiquities Josephus elsewhere consistently uses σημεῖον to mean a sign authenticating a prophet.

A stupendous miracle of liberation would also serve to authenticate the prophet who predicted it. This does not mean, however, that the signs need have been restricted to a purely authenticating function. Perhaps the ‘impostors’ hoped that Felix would send out a substantial force in pursuit so that Yahweh could dispose of them as he had once disposed of the pursuing Egyptians. This would then be not only an unmistakable sign that Yahweh was once again about to intervene on Israel’s behalf, but also the first instalment of that decisive intervention.

Josephus’s description of these unnamed figures under Felix is studiously vague, so no conclusion about them can approach certainty. Based on the evidence Josephus supplies, however, it may well be that they saw themselves as called to participate in the first-century equivalent of the great saving events of the exodus.

d. The Samaritan under Pilate

According to Ant. 18.85–87,

The Samaritan nation too was not exempt from disturbance. For a man who made light of mendacity and in all his designs catered to the mob, rallied them, bidding them go in a body with him to Mount Gerizim, which in their belief is the most sacred of mountains. He assured them that on their arrival he would show them the sacred vessels which were buried there, where Moses had deposited them. His hearers, viewing this tale as plausible, appeared in arms. They posted themselves in a certain village named Tirathana, and, as they planned to climb the mountain in a great multitude, they welcomed to their ranks the new arrivals who kept coming. But before they could ascend, Pilate blocked their projected route up the mountain with a detachment of cavalry and heavy-armed infantry, who in an encounter with the firstcomers in the village slew some in a pitched battle and put the others to flight. Many prisoners were taken, of whom Pilate put to death the principal leaders and those who were most influential among the fugitives.

Josephus does not explicitly state that this Samaritan promised a sign; but neither does he explicitly say that of Theudas or the Egyptian. It could well be that the discovery of the sacred vessels buried by Moses would function well enough as a sign for the excited Samaritans. That large numbers flocked to Mount Gerizim in expectation of their discovery certainly suggests so. That they did so armed, and that Pilate felt it necessary to suppress the project so forcibly that he was recalled by Rome, suggests so even more strongly. Clearly, these Samaritans were not innocuous pilgrims trying to recover cultic objects of purely religious significance. Although Josephus does not spell out the implications, it would appear that these Samaritans supposed that the recovery of sacred vessels buried by Moses would be accompanied by the recovery of other features of the Mosaic age, such as Yahweh’s assistance in liberating his people.

Josephus does not attach any particular label to this Samaritan. He is called neither a ‘(false) prophet’ nor a ‘charlatan’, but merely ‘a man who made light of mendacity’. The description nevertheless effectively labels him as a ‘deceiver’, and so ranks him with the other deceivers. Moreover, it could well be that his claim to be able to locate the buried vessels was seen as a prophetic one, although this is less clear.

It is interesting that this passage runs counter to the tendency of Antiquities to appear less militaristic than War. The heavy infantry and cavalry, more often featured in War, here appear in Antiquities. Moreover, it is in this Antiquities account that Josephus explicitly says that the movement was armed.

e. The Impostor under Festus

According to Ant. 20.188,

Festus also sent a force of cavalry and infantry against the dupes of a certain impostor (τινος ἀνθρώπον γόητος) who had promised them salvation and rest from troubles, if they chose to follow him into the wilderness. The force which Festus dispatched destroyed both the deceiver himself and those who had followed him.

This brief account could almost be taken as the paradigm of the ‘charlatan story’ form, since it highlights the stereotypical features of many of the others. An impostor leads dupes into the wilderness; the Romans deploy ‘a force of cavalry and infantry’; result: Romans destroy dupes. At first sight this short account adds little to what we have already seen. Yet its very shortness makes recurrent features of Josephus’s charlatan stories stand out and so raises the suspicion that Josephus tended to conform his charlatan stories to a standard pattern.

On this occasion, however, it is not said that the impostor promised a sign; he simply promised ‘salvation and rest from troubles’. It is not clear whether this meant an undisturbed life in the desert or whether it is a coded way of referring to something more sinister (plotting to drive out the Romans, for instance).

f. The Prophet at the Fall of Jerusalem

The next of Josephus’s charlatans makes his appearance at the downfall of Jerusalem in 70 ce:

They [sc. the Romans] then proceeded to the one remaining portico of the outer court, on which the poor women and children of the populace and a mixed multitude had taken refuge, numbering six thousand. And before Caesar had come to any decision or given any orders to the officers concerning these people, the soldiers, carried away by rage, set fire to the portico from below; with the result that some were killed plunging out of the flames, others perished amidst them, and out of all that multitude not a soul escaped. They owed their destruction to a false prophet (ψευδοπροφήτης), who had on that day proclaimed to the people in the city that God commanded them to go up to the temple court, to receive there the tokens of their deliverance (τὰ σημεῖα τῆς σωτηρίας) (War 6.283–85).

In this story the pattern is noticeably different. No longer is there a band of hopeful dupes set upon by Roman troops forestalling a potential revolt; instead there is a huddle of desperate dupes cremated by victorious Roman troops in the heat of battle.

The occurrence of the phrase σημεῖα τῆς σωτηρίας is noteworthy here. Whatever Josephus meant by ‘sign’ in connexion with the other sign prophets, the word can hardly mean ‘authenticating sign’ here. This prophet’s followers do not have the time to watch the shadow go into reverse on Hezekiah’s sundial, as it were; they need to see the angel of the Lord slaughter 185,000 enemy soldiers straight away. It is possible that, as Betz suggests, the prophet appealed to the tradition of Zion’s invulnerability.21

g. Jonathan the Sicarius

The last of the sign prophets is also different. His story is longer and more complex than those of his predecessors, and takes place after the fall of Jerusalem. It also involves Josephus personally, a mixed blessing from our point of view, since while this suggests that Josephus will have been reliably informed, it also gives him a strong motive for even more bias than usual. The opening of Jonathan’s story, at War 7.437–41, conforms broadly to the standard type. A deceiver (this time described as an ‘arrant scoundrel’, πονηρότατος ἄνθρωπος) leads a parade of dupes into the desert where he promises to show them ‘signs and apparitions’ (σημεῖα καὶ φάσματα). A force of Roman cavalry and infantry go in pursuit and easily overpower the unfortunates, killing some and capturing others. There are a few unusual features. For one thing, this story takes place in Libya instead of Palestine. For another, Josephus explicitly states that the crowd was unarmed. And for another, Josephus refers to Jonathan’s escapade as an ‘exodus’. The word may be accidental, but it is more likely to have been chosen, probably sarcastically, to recall the activity of Moses. The choice of φάσματα in place of τέρατα may also be a piece of calculated sarcasm: when Moses led the people of Israel in the real exodus God worked real signs and wonders through him for their salvation; but when Jonathan led his phoney exodus the most one could expect from him would be (phoney) signs and (mere) apparitions. The final difference is that, unlike the sign prophets before him, once Jonathan is caught he tries to save his own neck at the cost of others. Thus the story continues:

These calumnies were readily entertained by Catullus, who invested the affair with serious importance, pompously exaggerating it, in order than he too might be thought to have won a Jewish war. But—what was far worse—not only did he show this easy credulity, but he actually prompted the Sicarii in falsehood. Thus he instructed Jonathan to name one Alexander, a Jew, with whom he had formerly quarrelled and was now at open enmity, further implicating his wife Berenice in the allegations. These were his first victims. After them he slew all the well-to-do Jews, three thousand persons in all; a step which he thought that he could safely take, as he confiscated their property to the imperial exchequer (War 7.442–46).

Catullus’s villainy does not end there, for he goes on to bring charges of sedition against notable Jews elsewhere, including Josephus. Catullus then takes Jonathan to Rome to obtain a hearing, but Vespasian, on investigating the facts, finds Jonathan to be a liar. Jonathan is duly tortured and burnt alive, though Catullus escapes with a reprimand from the emperor. God proves less lenient, since Catullus is soon after smitten with a dire disease which, after providing a suitable amount of torment, eventually proves fatal (War 7.447–53; cf. Life 424–25).

Clearly Josephus felt no cause to love either Jonathan or Catullus. His account of their plotting and downfall was written with as much vitriol as he could muster. This does not prove that Josephus was guilty of systematic mendacity in constructing this account, but one may nevertheless observe how it performs its rhetorical function. To begin with, the archfiend Jonathan is identified as belonging to two groups Josephus’s writings have by now discredited: the sicarii and the deceiving sign-prophets. The way Josephus tells the story makes it clear that Jonathan is not even a genuine charlatan; unlike the others who may have succeeded in duping themselves and generally perished alongside the people they duped, Jonathan appears as an unscrupulous survivor, prepared to say anything that might get him out of trouble. Finding an eminently corruptible governor in Catullus, he then proceeds to pander to Catullus’s corruption. While one might suppose the other sign prophets to have been genuinely deluded, Jonathan is not even granted the title ‘false prophet’. By the time Josephus has finished with him, he is a figure beneath the contempt conveyed by such a title.

But then one has to ask whether Jonathan ever set out to be a prophet in the first place. The only things which make him appear so are the (possibly sarcastic) reference to ‘signs and apparitions’, the (possibly sarcastic) use of the word ‘exodus’, and the conformation of Josephus’s account of him to the general charlatan story form. Life 424 described Jonathan not as a prophet but as ‘a certain Jew, named Jonathan, who had promoted an insurrection in Cyrene’. Moreover, his specific accusation against Josephus was that Josephus ‘had provided him with arms and money’. On the face of it, someone who wants arms and money is planning armed rebellion, not waiting for God’s intervention. Given that War states that Jonathan was a sicarius and that the Life refers to him as an insurrectionist pure and simple, it seems likely that he was an insurrectionist pure and simple. Jonathan’s place among the sign prophets may be due solely to Josephus’s rhetoric.

Characteristics of the Sign Prophets

‘Sign Prophet’ is a modern scholarly term used to designate a number of different figures who appear in Josephus’s writings. Josephus may have made them appear more alike than they actually were. Gray suggests that Theudas and the Egyptian stand out as promising ‘epiphanies’ (i.e. saving acts of God) whereas the others only promised ‘signs’ (i.e. proofs that they were prophets sent by God). I have just suggested that Jonathan the Sicarius may have been not a sign prophet but an insurrectionist.23 Under Felix there appears to have been a group of sign prophets working together, whereas the others appear as individuals. The prophet at the fall of Jerusalem was responding to an emergency inside a city faced with imminent defeat, whereas most of the others operated outside the city without any apparently pressing emergency in the hope that God would shortly bring victory.

Nonetheless, there appear to have been some common features in this group, and it will be worth while to examine them, at the same time comparing the sign prophets with other figures such as Jesus and John the Baptist. Since there is no point in re-inventing a well-crafted wheel, this survey may conveniently be conducted under the list of characteristics proposed by Gray, plus two further ones.

a. Leaders of Sizeable Popular Movements

Gray acknowledges that ‘sizeable’ is deliberately vague, but suggests that the sign prophets were all leaders of relatively large bands who followed them from one place to another, unlike ‘popular prophets’ such as Jesus and John the Baptist. Jesus and John admittedly attracted large crowds while they were ministering in any one place, but were followed from one place to another only by a small band of disciples.

Although this seems generally correct, one might enter one or two reservations. On the one hand, Josephus could well be exaggerating the numbers of people involved in the sign prophet movements (see, for example, the above discussion of the number of people who followed the Egyptian). On the other, the Gospel accounts suggest that Jesus’ followers were not restricted solely to the Twelve. Perhaps the maximum number of itinerant followers Jesus had would still be smaller than the minimum number that the least successful of Josephus’s sign prophets attracted, but the difference may be more one of degree than of kind.

Gray’s second point is that in contrast to Jesus and John, the sign prophets ‘led sizable groups of people from one place to another in anticipation of some dramatic act of deliverance’. Yet whatever may have been in Jesus’ mind, the Gospels suggest in more that one place that his followers expected something dramatic to happen (specifically, the arrival of the kingdom of God) when they reached Jerusalem (e.g. Mk 10:33–37; Lk. 19:11). John’s notice of the crowd who followed Jesus after the feeding of the five thousand (Jn 6:24) is also noteworthy in this regard, especially given that their response to the feeding miracle was to declare Jesus a prophet and attempt to make him king (Jn 6:14–15).

b. The Movements were Drawn from the Common People

On this Gray comments further:

Some of the movements I have been considering seem to have arisen in Jerusalem … or in other towns (Jonathan); the rest may have formed in the countryside. There is little evidence that animosity between the rural population and town-dwellers played a significant role in the formation of these groups …

The followers of the sign prophets, then, were mostly ordinary people from the towns and countryside of Jewish Palestine … It is reasonable to suppose that any social or economic pressures bearing on the common people in this period would have contributed to the rise of movements, like those led by the sign prophets, which sought liberation or deliverance from the existing structures of society.

This seems fair enough, though having groups originate in Jerusalem, other towns and the countryside would seem to cover most of the available possibilities. Josephus does not give any information concerning the socio-economic condition of the sign prophets’ followers. He tends to refer to them somewhat contemptuously as the ‘mob’ or ‘multitude’, which does suggest that they came from the common people. Gray’s remarks represent reasonable suppositions based on the available evidence. One might add that, so far as one can tell, Jesus’ followers were drawn from similar kinds of people.

c. These Figures Presented Themselves as Prophets

Josephus explicitly states that Theudas claimed to be a prophet, and applies the term ‘false prophet’ to the Egyptian and the prophet at the fall of Jerusalem, which presumably indicates that they also claimed to be prophets. The collective figures who appeared under Felix are variously described as πλάνοι, ἀπατεῶνες and γόητες claiming to operate under divine inspiration. Festus’s impostor is likewise called a γόης. The Samaritan suppressed by Pilate is described merely as a man who made light of mendacity. Jonathan, who may not have been a sign prophet at all, is termed ‘an arrant scoundrel’. The term ‘prophet’ or ‘false prophet’ is thus applied explicitly only to three out of the seven individuals or groups mentioned. One must therefore ask whether Gray is justified in asserting ‘it is clear that they presented themselves as prophets and were regarded as prophets by their followers’. Gray herself points out that there were other popular figures around who claimed to be kings. She justifiably protests against lumping all these popular figures together as ‘messianic prophets’, but also observes that the would-be king and the would-be prophet ‘overlap to some extent in the person of the Egyptian’. The situation is further complicated by the possibility that a figure such as Moses, on whom some of these people seem to have modelled themselves, could be seen as a prophet-king.30 Given that there also seems to have been some expectation of a prophet like Moses, it may well have been that the boundaries between ‘messiah’, ‘prophet’ and ‘king’ were not all that clearly drawn.

Gray nevertheless proposes a criterion by which a line might be drawn. The messianic pretenders or would-be kings reported by Josephus appear to have had a reasonably practical political programme. They generally armed their followers, and sometimes managed to control an area of the countryside for a while. The prophetic figures, on the other hand, were not so concerned about political or military realities, since they were looking for a decisive intervention from God. This is clearly a good working distinction, and there may well have been persons who fell clearly on one side or other of the line it draws. But there may have been others who straddled it, as Gray thinks the Egyptian did.

How far can one decide on which side of Gray’s line each of the figures Josephus describes in fact falls? Theudas is termed both ‘prophet’ and ‘charlatan’. So far as one can tell from Josephus he relied purely on divine intervention; he expected God to part the Jordan, and he led what sounds more like a civilian exodus than a military campaign. The Egyptian false prophet is harder to place; according to Antiquities he was relying on divine intervention (to make the walls of Jerusalem collapse), whereas in the War account he was apparently hoping to take the city by storm. Even allowing for the probability that Antiquities is nearer the truth than War, this prophet seems not to have been wholly devoid of military ambitions. Whatever help he expected from God in gaining access to the city, he still had to occupy it and overcome its defenders, even if he was counting on their being thoroughly demoralized by the miraculous collapse of the city wall. The prophet at the fall of Jerusalem seems to have entertained no military ambitions of his own; he simply declared that God would act.

One thing these three figures (whom Josephus explicitly designates as [false] prophets) have in common is that they all promised that something spectacular would happen. Josephus elsewhere stresses the predictive element in prophecy. It thus seems reasonable that Josephus would allow the title ‘prophet’ or ‘false prophet’ to anyone who offered such a prediction, and that this would not have been out of keeping with the usage at the time.

A further thing the three figures may have in common is that their predictions are based on great saving acts from Israel’s past. The Egyptian almost certainly based his prediction on Joshua’s conquest of Jericho. Theudas was apparently hoping to recreate either the Red Sea crossing or Joshua’s entry into the Promised Land (perhaps with a hint of Elijah and Elisha thrown in for good measure). It is less clear what precisely the Jerusalem prophet was expecting, but he may have been hoping for something like the destruction visited upon Sennacherib’s army. The term ‘prophet’ may thus have had connotations of a figure who resembled one of the great men of God of the biblical past (cf. Mk 6:15; 8:28), someone like Moses, Elijah, Isaiah or even (stretching the term ‘prophet’) Joshua.

Armed with these two criteria, we may now determine which of the other figures Josephus mentions could reasonably be called ‘prophets’ even though Josephus does not explicitly use the term.

The Samaritan would seem to qualify. He makes a specific prediction (that vessels buried by Moses would be found), and judging by the popular reaction this was felt to be spectacular. Quite what the significance of this find was meant to be is unclear, but it was associated with a great man of God from the biblical past (Moses). One assumes that the Samaritan had not just equipped himself with a metal-detector or even a treasure-map, so the sacred vessels were presumably meant to be found through divine aid. On the other hand, it is interesting that the Samaritan attracted armed supporters. This gives the impression that the finding of the buried vessels would be a signal for a holy war. It could be, then, that this would-be prophetic Samaritan hoped to take on some of the characteristics of Moses, Joshua or the Judges.

The report of the impostor under Festus lacks sufficient detail to classify the man with any certainty. He is said to have promised salvation and rest from troubles. It is unclear whether this is a prophetic prediction or a political programme. It is equally unclear whether he led people out into the wilderness to witness miracles of salvation or to construct a base for revolutionary operations. The promise of ‘rest from troubles’ (παῦλαν κακῶν) could be a reference to the exodus-conquest tradition (Exod. 33:14; Deut. 3:20; 12:9, 10; 25:19; Josh. 1:13, 15; 11:23; 14:15; 21:44; 22:4; 23:1; Ps. 95:11), but this is highly uncertain. This impostor may or may not have claimed to be a prophet.

I have already suggested that Jonathan was not a ‘sign prophet’ but an insurrectionist. That leaves the group of ‘deceivers’ who collectively led an expedition into the desert at the time of Felix. These promised ‘wonders and signs’ (τέρατα καὶ σημεῖα) or ‘signs of liberation’ (σημεῖα ἐλευθερίας). As I have argued, both phrases could be taken as pointing back to the exodus tradition. Moreover, these were clearly figures who predicted that something spectacular was about to occur. They should thus probably be regarded as prophetic figures even though Josephus does not actually say so. They may have hoped to become militant prophets in the exodus-conquest tradition.

In summary, it appears that Theudas, the Egyptian, the Jerusalem prophet, the Samaritan and the group of deceivers under Felix all qualify as prophets or would-be prophets. Jonathan the Sicarius probably does not. The charlatan under Festus may or may not qualify. Some of these prophets may also have entertained more military, royal, or political aspirations as well. Both the Egyptian and the Samaritan would seem to come into this category. So might the characters under Felix and Festus.

d. Leading People to a Specific Destination

Theudas led his followers to the Jordan; the Egyptian led his via a circuitous route to the Mount of Olives. The Samaritan led his followers to Mount Gerizim, though Josephus indicates that supporters continued to flock there separately in order to join him. The impostors under Felix and Festus led their followers into the desert, as did Jonathan. The Jerusalem prophet led a band of people to the temple, but this is different. Jerusalem was under attack at the time; there were doubtless crowds of people eager to flee in any direction that suggested itself; the prophet appears to have hoped that the temple would provide a safe refuge. In this case the prophet’s followers were attempting to flee from imminent danger; in the others they were going out in search of adventure.

There are several different destinations here, and as Gray observes, the significance of the desert may not be the same in each case. It remains true that leading a band of people to a specific destination is a common characteristic of all these prophets. In contrast, people came out to a specific place (the Jordan) to be baptized by John, but he did not lead them there, and afterwards they returned to their homes (presumably this is what Mk 1:5 means; cf. Ant. 18.116–19).

Jesus of Nazareth is an intermediate case. He is certainly represented as calling people to leave their homes and follow him (e.g. Mk 1:16–20; 2:13–14). For much of the time, however, this following seems to have taken the form of an itinerant ministry round Galilee (e.g. Mk 1:39) rather than a movement towards any specific place. On one or two occasions Jesus is said to have fed a large crowd of people in the wilderness, though according to Mk 6:33 they followed him there against his will. Mk 8:1 merely states that a large crowd had gathered; Jesus goes on to say that they had been with him three days but it is not stated whether he had led them there or not. According to Jn 6:2 the multitude followed Jesus to a mountain because they saw the signs he performed on the sick—does this constitute leading a band of followers or not? Felix or Festus may well have supposed so had this occurred under their jurisdiction. All four Evangelists are agreed that towards the end of his ministry Jesus led his band of followers up to Jerusalem. If Jesus led his disciples to Jerusalem by a circuitous route, this is roughly what Josephus says of the Egyptian.35 Presumably the Egyptian differed from Jesus in the size of his following, and he appears not have taken the precaution of advancing upon Jerusalem at a major festival at which crowds of pilgrims would be expected in any case.

e. A Soteriological Message

Gray is correctly reserved towards labelling the sign prophets ‘eschatological’. This, she points out, may mean one of two things: either that these figures modelled themselves on some great figure of the past, such as Moses, whose coming (or the coming of one like whom) was expected at the eschaton, or that they proclaimed the coming of a new age, or some other radical transformation of the current world order to be brought about by a dramatic divine intervention. Against this, Gray maintains that it is unclear whether all the sign prophets should in fact be associated with a figure like Moses, or whether their message was always strictly eschatological in this sense. The Egyptian, for example, hoped to replace the existing authorities in Jerusalem; his goal was thus political revolution, not the inauguration of the age to come. Again, the prophet at the fall of Jerusalem was seeking miraculous deliverance from imminent danger, but there was no evidence that he associated this with the close of the present age.

These points are well made; it is not clear that these prophets were ‘eschatological’ prophets. But if one puts aside the Christian meaning of ‘soteriology’ then it may be said that most, if not all, of Josephus’s sign prophets had a soteriological message. As Gray puts it, ‘It may be said with a fair degree of confidence that the sign prophets announced to their followers that God was about to act to deliver them.’

In some cases, this is absolutely clear. The prophet operating at the time of Festus explicitly promised σωτηρίαν. The Jerusalem prophet likewise promised that those who followed him to the temple would see τὰ σημεῖα τῆς σωτηρίας. The prophetic group suppressed by Felix promised σημεῖα ἐλευθερίας, which also suggests something soteriological. Thereafter it becomes less clear. Theudas promised to part the Jordan and the Samaritan promised to discover sacred vessels buried by Moses. These could be taken as soteriological insofar as they recall and thereby hope to re-enact significant moments in Israel’s salvation history. Jonathan is said to have promised only a display of signs and apparitions, but then perhaps he should not be ranked among the prophets at all. If Josephus’s estimate of the Egyptian be taken at face value, then he was simply promoting his own political ambitions. It may be, however, that he was trying to emulate Joshua and that he did have a message of deliverance from the Roman (and maybe the existing Jewish) authorities.

John the Baptist is represented as a prophet of repentance, with a message more of judgment than salvation, though salvation is presumably implied for those who repent. Jesus preached the coming of the Kingdom of God, which was a soteriological concept.

f. The Promise of Miracles

According to Gray, ‘In connection with their announcement of imminent divine deliverance, the sign prophets promised their followers that God, or they themselves, would perform some sort of miracle.’ This is certainly true of many of them. Theudas promised to part the Jordan. The Egyptian apparently promised the collapse of the walls of Jerusalem. The Jerusalem prophet promised a miraculous divine intervention just as Jerusalem was falling. The prophets under Festus promised ‘wonders and signs’ or ‘signs of liberation’ in the desert.

In other cases it is not clear, however, that any miracle was promised. Josephus’s reference to the ‘signs and apparitions’ promised by Jonathan may simply be sarcastic. It is possible that the Egyptian was prophesying the downfall of Jerusalem in the manner of Jeremiah rather than setting out to conquer it in the manner of Joshua (though, to be sure, one must then account for his sizeable band of followers). The Samaritan promised to discover buried treasure. He may well have proposed to do so by miraculous means, and its discovery would presumably have been seen as a sign, but whether this is strictly the promise of a miracle is not so clear. Festus’s deceiver is said only to have promised salvation and rest from troubles.

Gray spends some time developing Betz’s distinction between sign and epiphany in Josephus, and arguing that for Josephus, σημεῖον has the special sense of ‘sign used to authenticate a prophet as a true messenger of God’. In Chapter 2 I argued that, apart from one or two exceptions, Josephus does indeed restrict σημεῖον to this sense throughout most of Ant. 1–10. From Ant. 10.34 onwards, however, this special meaning is dropped, and at Ant. 10.234; 18.211; and 19.9, 94 the word means ‘omen’ or ‘portent’ (as it also tends to mean in War when it refers to anything miraculous). Gray attempts to argue from the special meaning of σημεῖον in the first half of Antiquities that Josephus also intends the word in this sense when he uses it in connexion with the sign prophets. Noting that Josephus frequently uses σημεῖον to mean ‘omen’ or ‘portent’ she further suggests that omens and authenticating signs perform similar functions for Josephus, except that the latter are announced in advance whereas the former simply happen.

Part of the problem with Gray’s argument is that Josephus never refers to what the sign prophets promise simply as a ‘sign’. The word σημεῖον is not used at all in connexion with Theudas, the Egyptian, the Samaritan or the impostor under Festus. The prophets at the time of Felix promise either σημεῖα ἐλευθερίας or τέρατα καὶ σημεῖα. The prophet at the fall of Jerusalem promises τὰ σημεῖα τῆς σωτηρίας. Jonathan the Sicarius leads people out to see σημεῖα καὶ φάσματα in the wilderness. Gray is forced to concede that ‘it is unusual for Josephus to use the terms τέρας and φάσμα to designate authenticating signs, as he apparently does in War 7.438 and Antiquities 20.168′. But this is an admission that threatens to undermine half the cases in which the word σημεῖον is used in connexion with a sign prophet. One could argue that it undermines two-thirds of the cases, since the problematic τέρατα καὶ σημεῖα of Ant. 20.168 presumably refer to the same phenomena as the σημεῖα ἐλευθερίσς of War 2.259. This leaves the Jerusalem prophet with his σημεῖα τῆς σωτηρίας. These cannot possibly be signs the principal purpose of which is to authenticate him as a prophet, as I have already argued; people with bloodthirsty soldiers on their heels would have more pressing concerns.

Wherever Josephus qualifies σημεῖον with a genitive, another noun joined by καί, or some other such expression, it is unlikely to mean simply ‘authenticating sign’. The most glaring exception to his special usage in the first half of the Antiquities occurs at Ant. 2.327 where the Israelites are forgetful of signs already wrought πρὸς τὴν ἐλευθερίαν. The qualification ‘for liberation’ here would seem to exclude the possibility that the primary purpose of these particular signs was ‘for authentication’ (a point I have again already argued). The similar phrase σημεῖα ἐλευθερίας at War 2.259 is thus likely to have a broadly similar meaning. By the time they reached the Red Sea the Israelites had already witnessed signs that had worked towards their liberation (from Egypt). The prophets under Felix may have been looking for signs of imminent liberation from the Roman Empire. The Jerusalem prophet presumably promised signs of deliverance from the approaching legionaries. It may be that such signs would also authenticate the prophet who promised them, and it may even be that persons claiming to be prophets would be asked for such signs, but the primary purpose envisaged for these signs was probably to assure the witnesses that God was acting or about to act on their behalf.

John the Baptist is nowhere recorded as having performed a miracle, though some scholars feel Jn 10:41 protests too much. The Synoptic Gospels famously report Jesus refusing to perform a sign, although in the context in which Mark places the refusal (almost directly after the feeding of the four thousand) it may well be ironical. If Gray is wrong in thinking that the sign prophets promised signs that were primarily meant to authenticate their own ministries, then in any case the distinction with Jesus becomes less clear. Jesus may well have intended his miracles, at least in part, as signs of the coming of the kingdom or of the nature of the kingdom. They most likely conveyed a soteriological message whether or not they would also be understood as eschatological; that is, they could have been signs of what God was doing or about to do through the ministry of Jesus whether or not people connected them with the eschaton. This also seems to have been the case with many of the signs promised by Josephus’s sign prophets; they were signs of what God was about to do for his people. Where Jesus differed from the sign prophets was in his understanding of what God was doing.

A further important distinction between Jesus and the sign prophets is that whereas Jesus apparently actually did perform a number of miracles (healings and exorcisms at least), there is no indication that the sign prophets in fact did so (and healings and exorcisms seem not have formed part of their repertoire). To be sure, Josephus might have had his polemical reasons for suppressing the fact that the sign prophets were able to deliver on their promises, but it seems most unlikely that they were. Had the Jordan in fact parted for Theudas and the walls of Jerusalem collapsed for the Egyptian, the course of Jewish history would have run rather differently.

g. Suppression by Roman Authorities

A feature common to all Josephus’s sign prophet reports is the ignominious end of the prophetic movements in the face of Roman arms. Fadus suppressed Theudas with a squadron of cavalry. Felix crushed the Egyptian’s followers with infantry, and possibly cavalry as well. Forces of cavalry and infantry also figure in the demise of the movements led by the collective charlatans under Felix, the Samaritan under Pilate, and Jonathan the Sicarius. Those led up to the temple by the Jerusalem prophet likewise perished by Roman arms, but their case was different: they were simply in the way of a victorious army at the height of battle. In each of the other cases, however, a Roman official took a deliberate decision to employ Roman arms to suppress what he clearly saw as a political threat.

Gray argues that the sign prophets were not political in the sense of having ‘a practical plan for ousting the Romans by force’. Instead, they ‘believed that the deliverance they expected and announced would be wrought miraculously by God; it would not be achieved through their own efforts alone’. She further notes that Jesus and John the Baptist were ‘apolitical’ in this sense, though they too were executed by the authorities. The question then arises why the followers of Jesus and John were apparently allowed to escape unmolested, whereas the Romans troops struck heavily against the followers of the sign prophets. Gray suggests that

it is possible that the Romans responded differently to the sign prophets not because they believed that they were any more political than John or Jesus, but because their followers were much more numerous than in the case of John and Jesus, and were all gathered together in one spot, making them, at one and the same time, a more obvious threat to public order and an easier target for military action.

This seems an eminently sensible suggestion, but there are others one might make. For one thing, John was not executed by the Romans but by Herod Antipas. Josephus’s account states that Herod saw the drawing of crowds as a threatening activity in itself. The Romans may well have thought the same, but John was apparently not operating in territory under direct Roman jurisdiction. The same is true for Jesus throughout most of his ministry. Thus Smith remarks:

The crowds following him were not broken up by military force because he was not in Roman territory but in that of Herod Antipas, and the Roman government frowned on the use of violence by its subordinate rulers against their subjects; Antipas’ younger brother had been deposed for that, and the lesson had not been lost.

Whether it is true that the Romans generally frowned upon subordinate rulers employing violence to keep order may be doubted; there is little to suggest that Herod the Great’s violence troubled them overmuch. But the point nevertheless remains that Herod Antipas’s policy may have differed from that of Rome.

Although part of the reason why the followers of John and Jesus received different treatment from the Romans from that accorded to the followers of the sign prophets may thus be the difference in whose jurisdiction they operated, the fact remains that Jesus was executed by the Roman authorities after leading his followers into territory that was under direct Roman rule. Yet, several factors may have been operating that distinguished him from the sign prophets. First, as Gray suggests, his following may simply have been smaller. Secondly, if Jesus went up to Jerusalem at Passover, his band of followers would probably not have looked any different from the other bands of pilgrims going up for the festival. Thirdly, in the other instances, troop deployments took place in the open countryside; but to have attempted to deploy cavalry and heavy infantry in Jerusalem in the middle of a major festival in order to suppress one relatively small prophetic movement may have struck even Pilate as unwise. In this case, it would have seemed easier to deal with the leader and hope that the problem would then go away, especially if (as the Gospels indicate), his followers seemed more intent on running away than standing by him. Finally, in all the instances reported by Josephus, the Roman authorities would presumably have perceived the sign prophet as leading a sizeable band of persons from one place to another. The fact that Jesus entered Jerusalem at Passover time, and may have been quite unknown to the Romans beforehand, could well mean that he was not perceived by them in the same light. Indeed (again as the Gospels indicate), it may well have been the Jewish temple authorities that were instrumental in persuading Pilate to take action, and this may also explain the difference in approach.

h. Operation in Jewish Palestine

If one excludes Jonathan as a dubious case, then all Josephus’s sign prophets, in common with Jesus and John, operated in Jewish Palestine (and mainly in Judaea). The point is almost a trivial one but for two things. First, all these people were operating within a society whose expectations and beliefs were shaped by various Jewish traditions. It therefore becomes more plausible to see a connexion with, say, Moses and exodus traditions than it would if similar events took place among pagans in Italy or Asia Minor, say. Secondly, and conversely, Josephus’s reports about the sign prophets provide an indication of what the popular mood in Jewish Palestine was around this time. Josephus himself is utterly contemptuous of the rabble who so readily allowed themselves to be duped, and it is very easy for the modern reader, who naturally tends to regard the sign prophets’ promises as absurd, to concur without much reflection. It might be more helpful to see these reports as an index of the mixture of hope, desperation and fervour that must have been around at the time for these would-be prophets to flourish to the extent they did. Not long after the time of Jesus, and hence probably during his time too, there were apparently significant numbers of people willing to believe the claim that someone was a prophet who would perform one or more signs to show that God was about to intervene dramatically in the affairs of his people.

i. Summary

‘Sign prophet’ is an artificial term. Josephus does not call all these people (false) prophets, nor does he always say explicitly that they promised signs. Moreover, one should not let his tendency to describe them according to a set pattern disguise the fact that they may have been a fairly disparate group. What one can say is that there do seem to have been a number of figures, popularly perceived as prophets, who promised one or more signs and who succeeded in attracting substantial followings on the basis of that promise. Where these movements took place within Roman jurisdiction, the authorities invariably repressed them by force of arms. At the very least, then, they were seen as a threat to public order; some of them may have been seen as potential revolutions. Precisely what the sign prophets believed or promised would happen is unclear, but it appears likely that at least some of them were looking for a divine intervention in the manner of the exodus and conquest traditions. It is very likely that the popular mood reflected in these movements was already current at the time Jesus and John ministered. Nevertheless, all the sign prophets Josephus records are later than Jesus.50

So far as one can tell, the sign prophets and those they led were looking back to the exodus—conquest traditions as the type of salvation and liberation expected in the imminent future. If so, then their interest in miracle was not unlike that reflected in much of the Jewish literature we have surveyed, which also shows a tendency to focus on the miracles of the exodus period and to celebrate Yahweh’s miraculous interventions on behalf of his people in battle. Even if Jesus did promise the ‘sign’ of a miraculous new temple, this would be tapping into a different stream of expectation. The feeding and sea miracles attributed to Jesus contain some echoes of the exodus tradition, but I have already argued that the resemblances are superficial. Even if Jesus did symbolically recreate the wilderness feeding, this was a miracle of God’s bountiful provision, not of punishment and defeat inflicted on Israel’s foes, and in any case the story as narrated makes Jesus resemble Elisha far more closely than Moses.

The fact remains that the most characteristic miracles of the historical Jesus were healings and exorcisms, types of miracle that not only receive comparatively scant attention in Second Temple Jewish literature, but seem to have formed no part of the programme of the sign prophets. There is, however, some evidence that other Jews around the time of Jesus practised exorcism, and the next chapter will go on to examine that evidence.

Chapter 12

Jewish Exorcists


We have found few stories in pre-Christian Jewish literature of demons being cast out of (as opposed to being driven away from) people. Even in the first century, accounts of Jewish exorcisms are rare outside the New Testament, the example most commonly quoted being Josephus’s story of Eleazar. It is, in fact, the New Testament, more than any other contemporary document, that creates the impression that Jewish exorcists were relatively commonplace. This chapter will review the literary evidence, both inside and outside the New Testament, to gauge how prevalent Jewish exorcists may actually have been. The following chapter will explore some anthropological perspectives on spirit possession in an attempt to correct the elite bias that may result from depending solely on literary evidence.

First, however it is necessary to clarify what is meant by exorcism. As we have already seen, the term is sometimes applied to just about any kind of procedure employed against demons, just as the term ‘possession’ is sometimes used for any kind of demonic attack. For the purposes of delineating the context of Jesus’ miracles, this usage is too broad. That is not to say that a discussion of demonology and apotropaic measures in general has no place in such an investigation; clearly it does; but notwithstanding the aetiology of the illnesses Jesus cured (on which see the next chapter), his exorcisms appear to form a distinct group of miracles overcoming an equally distinct form of possession. The concern of the present chapter is with other Jewish exorcists who might be comparable with Jesus. For that purpose it is necessary to concentrate on possession in the narrow sense of a demon invading a human victim and taking over control, and exorcism in the equally narrow sense of a reversal of this form of possession.

This is not to deny that there also existed a broader concept of ‘possession’ in which the demon invaded the sufferer’s body to bring about disease, and a corresponding concept of exorcism to bring about the cure of the disease. The distinction proposed is that between demonic ‘possession’, conceived as the invasion of the psyche resulting in the appearance of a different persona, and demonic ‘affliction’, more broadly conceived as any kind of demonic attack. This is precisely the distinction made by Meier between demonic possession and demonic obsession. Clearly this is a perfectly legitimate distinction for a modern writer to make, but it may be questioned whether it would have been a significant distinction for the authors of the ancient texts under consideration. This is a question that can be considered only briefly here, but the following eight considerations taken together suggest that this is a distinction that would have been made in the first century.

(1)    Cultural anthropological work on spirit possession suggests that this is a widespread phenomenon in which an altered state of consciousness is culturally interpreted as due to the influence of an alien spirit.

(2)    Structural-functional models of spirit possession indicate that the change in the subject’s persona, which facilitates a change in the relationships between the possessed person and those around him or her, is central to an explanation of its social function.

(3)    Although it is perfectly true that at least some cultures perceive a continuum between various forms of demonic attack, so that, for example, paralysis or stomach pains may be seen as an early stage in an attack leading to full-blown possession, the fact that these cultures distinguish an earlier and later stage shows that they are also distinguishing the phenomena in question. This is further supported by the fact that in the case of an illness such as paralysis or stomach pains, the onset of demonic possession is normally only diagnosed after alternative explanations, such as sorcery or the evil eye, have been eliminated. The fact that several types of suffering might be attributed to the same cause (i.e. both possession and other forms of affliction to demonic attack) does not mean that no distinction is made between the forms of suffering: in our culture we attribute both AIDS and the common cold to ‘a virus’ while clearly distinguishing between the two diseases!

(4)    Several Old Testament passages may be cited to show that the phenomenon of spirit possession as a change in persona seems to have been known in Israel prior to the rise of an elaborate demonology (hence it is prima facie likely that the phenomenon would continue to be regard as a distinctive form of demonic attack). The Hebrew terminology employed in the Hebrew scriptures for possession appears to relate to ecstatic behaviour or to mediumistic types of divination whereas the discussion of 4Q560 by Penney and Wise, suggests that a characteristic Aramaic terminology for possession normally included expressions meaning ‘to enter into [the body]’ (in this instance עלל בבשרא), a notion consistent with (though not strictly demanding) the concept of possession espoused here.

(5)    Several New Testament scholars (e.g. Stevan Davies, Paul W. Hollenbach and J. Duncan M. Derrett) who apply cultural anthropological insights to possession and exorcism in the New Testament effectively presuppose the narrower definition of possession assumed here. For points 1–5 generally, see the discussion in Chapter 13 Section 4 below.

(6)    Three out of Mark’s four exorcism stories (Mk 1:21–28; 5:1–20; 9:14–29) describe a situation in which the possessing demon apparently controls the behaviour or speech of the possessed. In the fourth (Mk 7:24–30), no clue is given concerning the girl’s symptoms, unless the clause εὗρεν τὸ παιδίον βεβλημένον ἐπὶ τὴν κλίνην καὶ τὸ δαιμόνιον ἐξεληλυθός (7:30) be taken to imply that the demon ‘threw’ the girl (taking βεβλημένον in its strongest sense) onto the bed before departing, but in any case this story hardly provides a counter-example. Mark 1:34 makes a distinction between healing the sick and casting out demons, and states that Jesus would not permit the demons to speak; presumably the only way they could speak would be through the vocal chords of those they possessed (as appears to be envisaged at Mk 1:24). Mark 6:13 also lists casting out demons and healing as separate activities. Although Mk 3:11 could be interpreted otherwise, it is at least consistent with the view that it was the possessed who fell down and cried out, which would imply that the demons were perceived as controlling them. Note that in the Beelzebul controversy (Mk 3:20–29) Jesus is not, strictly speaking, charged with being possessed, but with having Beelzebul (3:22) or having an unclean spirit (3:29), which could mean, having such a spirit under his control (i.e. being a sorcerer or spirit controller, rather than a possessed person or spirit-medium; see Chapter 13 Section 4 for clarification of these terms). There is thus no clear indication that Mark failed to make the distinction I am proposing, and several indications that he did; the balance of probability is that he did so, which in itself would suffice to show that what I have termed ‘possession’ in the strict sense could be seen as a distinct phenomenon in the first century. This is in no way inconsistent with Mark’s having attributed attacks other than possession to demons as well (e.g. the storm on the lake, see Mk 4:39; see also Chapter 13, Section 4 below for a discussion of the possibility that Mark may have attributed all illnesses to demons, but nevertheless regarded possession as a special case).

(7)    Mark is not the only writer who appears to describe what looks like my narrow concept of ‘possession’ when describing possession or exorcism. At Acts 19:13–17 (on which see Section 2.c below) the evil spirit apparently speaks through the possessed person and provides him with exceptional strength (a phenomenon attested in the anthropological literature on possession). The spirit of divination cast out by Paul at Acts 16:16–18 likewise speaks through the slave girl it possesses, and hence Luke presumably envisaged it as having at least partial control of her mind. The Syrian exorcist described by Lucian at Philopseudes 16 (see Section 4 below) deals with a case of possession that looks very similar to those envisaged in Mark. When Philostratus describes ApoUonius of Tyana casting out at demon at Life of Apollonius 4.20, it is clear that the demon has been in control of the youth’s personality both from what Apollonius says and the change in personality that follows the exorcism. The demon dealt with by the Indian sage at Life of Apollonius 3.38 is said to speak through the boy it possesses, in a voice which is unlike the boy’s, which, in addition to what the boy’s mother says about her son’s dire condition, again indicates that the demon (in this case the ghost of a departed warrior) is in at least partial control of its victim’s mind and body. The fact that Apollonius deals rather differently with a rather differently conceived plague demon at 4.10 does not negate the fact that Philostratus has twice exhibited the narrow concept of ‘possession’ I am defending. It is admittedly unclear what were symptoms exhibited by the demoniac exorcised by Eleazar at Ant. 8.45–49 (on which see Section 3.a below), though the overturning of the bowl of water by the departing demon does suggest a parallel with the overturning of a statue by the departing demon in Life of Apollonius 4.20, and hence may indicate that the nature of the possession was also similar. In any case, the Eleazar story fails to provide a clear counter example to the point I am making. To anticipate the conclusions of the present chapter, storises about exorcisms are relatively rare. It is thus surely significant that the majority of exorcism stories that survive appear to envisage ‘possession’ in my narrow sense. This is in no way inconsistent with the possibility that demons may also have been believed to attack in other ways as well.

(8)    In general one should also distinguish between the demonology that can be established from the text and the phenomenology of possession suggested both by exorcism stories and cultural anthropology. The fact that the former attributes a wide range of human ills to demons does not negate the indications of a distinct concept of possession exhibited by the latter, even if the demonology of the mythological background was used to interpret worrisome behaviour interpreted as demonic possession in my narrow sense.

The distinct forms of ‘exorcism’ and ‘possession’ defended here will accordingly be presupposed in what follows, an evaluation of the evidence for Jewish exorcists roughly contemporary with Jesus.

New Testament Evidence

a. Jesus and Beelzebul

At Mk 3:22 Jesus is accused of being possessed by Beelzebul, subsequently identified as the ‘prince of demons’. The scribes’ accusation is that it is the diabolical power by which Jesus is possessed that enables him to cast demons out of other people. The logic of Jesus’ reply is that if Satan were ordering his own subordinates to withdraw from their conquests, his kingdom would be in a state of civil war and accordingly on the point of collapse. Jesus’ success shows that the strong man (presumably Satan) has been bound so that one may now plunder his possessions (i.e. liberate those occupied by his minions). Failure to recognize the true nature of this activity is the unforgivable sin against the Holy Spirit, since when confronted with the power of God at work, it sees only the power of evil.

Mark’s scribes may have supposed that Beelzebul was ordering his subordinates to make tactical withdrawals in order to accredit a false prophet. What is less clear is whether Mark’s scribes at least had the excuse that they were faced with something novel (a successful exorcist), or whether they were hypocritically condemning in Jesus a practice they commended in others. It may thus be a piece of typically Matthean tidying up when in his version Jesus is made to ask, ‘And if I cast out demons by Beelzebul, by whom do your sons cast them out?’ (Mt. 12:27a). To be sure, this saying is generally regarded as a piece of Q material. But those not committed to the two-source hypothesis may see the placing of this saying as Matthean redaction (copied by Luke), whatever its origin. For it certainly serves to plug a hole in Jesus’ reply. If the Pharisees’ own people regularly cast out demons, then they have no excuse for applying a double standard when it comes to Jesus. In a similar vein the following verse (Mt. 12:28), picking up the theme of Satan’s kingdom already found in Mark, makes explicit the idea that defeats for the kingdom of Satan are victories for the kingdom of God, so that it is God, not Satan, who is at work here.

But even if these verses flowed directly from Matthew’s pen without the aid of any prior tradition they would have little force unless Matthew’s Pharisaic opponents practised exorcism. If the synagogue across the street was arguing that Jesus was a magician, Matthew’s attempt to tie the loose end left by Mark would be of little value unless his audience were prepared to believe that persons sanctioned by the synagogue cast out demons as well. Thus, even if Mt. 12:27 is not evidence for Jewish exorcists in Jesus’ day (though it may be), it is at least prime facie evidence for their existence in Matthew’s day.

The spirit/finger of God saying at Mt. 12:28 and Lk. 11:20 provides a possible clue to the nature of these Jewish exorcists. In the context in which the saying appears, the implication is that the sons of the Pharisees cast out demons by some agency other than the spirit/finger of God, since it is most unlikely that either Jesus or the Evangelists intend to attach eschatological significance to Pharisaic exorcisms. In this saying Jesus goes beyond merely rebutting the charge that he casts out demons by a demonic agency, he turns the charge upon his accusers by implying that the source of their exorcistic power is illegitimate. Jesus claims to cast out demons by the power of God. In contrast his accusers must employ other means, if not diabolical power then merely human art, either way something far closer to sorcery than Jesus’ exorcisms are. This counter accusation is made even more explicit in the Lukan version, where ‘finger of God’ is presumably an allusion to Exod. 8:19; when the Egyptian magicians are unable to reproduce the plague of gnats they declare, ‘This is the finger of God.’ Although the Pharisees make no such declaration, the implication is that their exorcists, who do not serve the Kingdom of God, are likewise mere magicians.

Luke elsewhere represents Jewish exorcists (and other wonder-workers) as magicians (see below on the sons of Sceva), so it would be in character for him to emphasize the charge here. Although Luke exhibits a lively interest in the Spirit, in his Gospel this is mainly confined to the first four chapters; thereafter Luke largely keeps the Spirit in reserve until Pentecost. There is therefore no reason why he should not have altered ‘spirit’ to ‘finger’ in order to create a scriptural allusion. Matthew’s ‘spirit’ may thus be the more original form of the saying. But in that case Luke has merely made more explicit what is already implied in Matthew, since ‘spirit of God’ or ‘finger of God’ presumably mean much the same thing, namely ‘power of God’. So far as these two Evangelists are concerned, Jesus’ exorcisms are effected through the power of God and are therefore true miracles, whereas Pharisaic exorcisms rely on some other source of power and are therefore magical.7

That Matthew and Luke should wish to represent Pharisaic exorcists as magicians is hardly surprising. Whether there is any justice in the charge is another matter. In the end, it could all come down to the divergent theological perspectives of the two parties to the debate: God is with us, not you, so we work miracles whereas you perform magic. Moreover, since the placing of Mt. 12:28/Lk. 11:20 is almost certainly redactional (whether the redactor was Matthew or the compiler of Matthew’s source), we cannot be sure that it intended any reproach against Pharisaic exorcists in its original context of utterance; the original emphasis may, for example, have been purely on the imminence of the Kingdom. Nevertheless, three considerations lend some support to the use Matthew and Luke make of this saying. First, as we have already noted, Jesus is hardly likely to have intended to imply that everyone’s exorcisms exhibit the power of the Kingdom. In almost any plausible context of utterance in Jesus’ ministry, Mt. 12:28 implies that Jesus saw empowerment by God’s spirit as distinctive of his exorcisms: the apodosis of the saying (‘the kingdom of God is upon you’) is hardly an everyday occurrence, therefore the protasis (‘If it is by the spirit of God …’) must envisage something equally extraordinary. This carries the further implication that (at least in Jesus’ view) other exorcisms were not so empowered, even if the main thrust of the saying is directed elsewhere. Moreover, the very claim to be casting out demons by the Spirit of God is itself distinctive. Secondly, though of itself this is a weak point, that Matthew and Luke indirectly indicate that Pharisaic exorcisms were magical rather than miraculous suggests that such a view may have had some plausibility. What lends weight to this second point is the third, namely that both Josephus and to a lesser extent some rabbinic literature also create the impression that Jewish exorcists worked by magical means, as we shall see below.

b. The Strange Exorcist

Mark 9:38–39 contains a brief story concerning the so-called ‘strange exorcist’:

John said to him, ‘Teacher, we saw a man casting out demons in your name, and we forbade him, because he was not following us’. But Jesus said, ‘Do not forbid him; for no one who does a mighty work in my name will be able soon after to speak evil of me.’

This suggests that someone, presumably a non-Christian Jew, was successfully casting out demons using the name of Jesus. Some commentators concede that this situation could have arisen during the ministry of the earthly Jesus, but feel that the pericope is more likely to be the product of the early Church, partly on the grounds that had Jesus in fact ordered so tolerant a policy, the early Church would not have had such a problem with non Christian exorcists using Jesus’ name. The contrary view, namely that this pericope goes back to the ministry of the earthly Jesus, is argued by Twelftree.10 His arguments, however, fail to convince. For example, in defence of the phrase ἐν ὀνόματι being used in Jesus’ lifetime he cites Acts 19:13 to show that ‘the sons of Sceva are said to be very quick to pick up the name of Paul as a possible source of power-authority’. But this is not what Acts 19:13 actually says, since the precise formula said to be employed by the exorcists in question is ‘I adjure you by the Jesus whom Paul preaches.’ In Twelftree’s terminology, it is Jesus, not Paul, who is the power-authority here.

For our purposes, however, it does not greatly matter whether this story originated during the time of Jesus or shortly after his death. That non Christian exorcists used the name of Jesus is known to us from other sources (e.g. the Magical Papryi). What the present story suggests is that such non Christian exorcists were around somewhere between the time of Jesus and the time of Mark. This places them in the first century, which is as accurate as we need to be.

c. The Sons of Sceva

According to the account in Acts 19:13–20, following the extraordinary miracles that God performed even through Paul’s handkerchiefs, a group of itinerant Jewish exorcists attempted to cast out demons using the formula quoted above. Seven sons of a Jewish high priest named Sceva were doing this, but the demon refused to acknowledge their authority to use Jesus’ name. Instead, the demoniac leapt on the would be exorcists, forcing them to flee from the house naked and wounded. As a result of this reversal, Jesus’ name came to be held in great awe in Ephesus, so that a great number of magicians gave up their magic and burned their magic books, and a great number of people became believers.

There are several problems with this account. If Josephus’s list is accurate, no Jewish High Priest up to the fall of Jerusalem held the name of Sceva. The Greek Σκεῦα probably represents the Latin ‘Scaeva’, which would be an unlikely name for any Jewish priest. Moreover, it is unclear why the sons of a Jewish High Priest should be pursuing a career as wandering exorcists in Ephesus, or why they should operate as a team of seven (which hardly suggests that they possessed any great prowess as individuals). It is also strange that although seven exorcists attempt to drive out the demon, the demon drives out ‘both’ of them (ἀμφοτέρων, v. 16). A further puzzle is why the failure of a group of exorcists to cast out a demon in the name of Jesus should result in Jesus’ name being extolled (v. 17). The confusion is further compounded by variant readings, particularly those of the Western text, which appears to tell a different story in which the seven sons of Sceva are a group distinct from the itinerant Jewish exorcists.

It is possible that Luke used ἀμφοτέρων to mean ‘all’ rather than ‘both’, since this usage is attested elsewhere in late Greek. Since the precise number of exorcists taking part in this encounter is not pertinent to the purpose of the present enquiry, we need spend no further time on it.

Secondly, it is not clear that ἀρχιερεύς need be translated as (officially reigning) ‘High Priest’. The New Testament frequently uses the word in the plural although only one High Priest could have held office at any one time. When Luke refers to an individual as the High Priest, he calls him ὁ ἀρχιερεύς (Lk. 22:50, 54; Acts 4:6; 5:17, 21, 27; 7:1; 9:1; 22:5; 23:2, 4; 24:1), except in Paul’s speech at Acts 23:5. Elsewhere Luke refers to ‘chief priests’ ([οἱ] ἀρχιερεῖς) in the plural (Lk. 9:22; 19:47; 20:1, 19; 22:2, 4, 52, 66; 23:4, 10, 13; 24:20; Acts 4:23; 5:24; 9:14, 21; 22:30; 23:14; 25:2, 15; 26:10, 12), where they are usually conjoined with some combination of ‘scribes’, ‘elders’ and ‘leaders of the people’. In the plural, the ‘chief priests’ appear as the temple authorities. Conversely, when Luke does give the High Priest a name, it is Caiaphas, Annas or Ananias (Lk. 3:2; Acts 4:6; 23:2; 24:1). It is thus unlikely that Luke imagined that the reigning High Priest at the time of Paul’s stay in Ephesus was called Sceva, and at least possible that when, as at 19:14, he uses ἀρχιερεύς without the definite article he means to refer, not to the High Priest, but to a chief priest. That a Jewish chief priest should bear the name ‘Sceva’ is still problematic, but the lack of a High Priest of that name is not strictly an issue.

A number of suggestions have been made to account for this character. Several commentators suggest that he may have been the self-styled high priest of a pagan cult, trying to pass himself off as a Jew, or else regard the title as merely an ‘advertisement’, the ancient equivalent of styling oneself professor’ or ‘captain’. Another possibility is that Σκεῦα represents, not the Latin name ‘Scaeva’ but a Jewish name or Hebrew word such as ‘Sheba’. Many scholars suppose that Luke took Sceva to be a Jew, as the Ἰουδαίου of verse 14 (in many, but not all, manuscripts) implies. F.F. Bruce suggests that since Luke did not have the device of quotation marks, he was simply giving the man’s own account of himself (without necessarily agreeing with it).18 Supporters of the originality of the Western Text, however, argue that Sceva was originally intended by Luke to be a Gentile.

The majority text makes reasonably good sense as it stands. One might argue that Luke deliberately makes Sceva a chief priest in order to make his sons a suitable foil for Paul’s success; even the sons of a high-status Jew try to imitate Paul’s use of the name of Jesus, but at the same time prove unable to so successfully. Had these exorcists been merely rogues, Paul could have gained little honour at the expense of persons whose honour-rating was already so low. The difficulty here is that Luke’s narrative has Jesus’ name extolled as the result of a demon’s victory. The contrast is not directly between Paul’s success and the Jewish exorcists’ failure, since in this story no one drives the demon out.

An alternative is to read this passage with the focus on magical practices (Acts 19:19). The Jewish exorcists, of whom the sons of Sceva are particular exemplars, make the same mistake as Simon Magus by assuming that the Christian missionary is merely performing a powerful form of magic, in this case by calling on the name (and hence the spirit) of someone who had died a violent death to perform an act of power. But the demon knows that the sons of Sceva are not authorized servants of Jesus Christ, and that they therefore have no authority to use his name. Accordingly, the demon is able to trounce the would-be exorcists. At the same time, in saying that it knows Jesus and Paul, the demon is saying that it would have acknowledged their authority.

This leaves the problem of what makes the strange exorcist who uses Jesus’ name at Lk. 9:49–50 acceptable, while the sons of Sceva are lampooned. It may be that Luke does not understand ἐν τῷ ὀνόματί in Lk. 9:49 to be equivalent to ὀνομάζειν in Acts 19:13. The former represents the correct use of the name, the latter its misuse. The difference is between appealing to Jesus’ authority (‘in your name’) and attempting to coerce Jesus’ spirit (‘pronouncing the name’). According to Acts 19:13 the Jewish exorcists set out to ὀνομάζειν ἐπὶ τοὺς ἔχοντας τὰ πνεύματα τὰ πονηρὰ τὸ ὄνομα τοῦ κυρίου Ἰησοῦ λέγοντες, Ὁρκίζω ὑμᾶς τὸν Ἰησοῦν ὅν Παῦλος κηρύσσει. Nowhere else in the New Testament is anyone described as an ἐξορκιστής, and Jesus is never said to ‘adjure’ (ὁρκίζειν) even when casting out demons (whom he generally ‘commands’ or ‘rebukes’). The only time ‘adjure’ occurs in connexion with one of Jesus’ exorcisms is at Mk 5:7, when the demon attempts to adjure Jesus, which could well be an unsuccessful attempt at self-defence by magical means.

There remains the question of the Western text. According to the majority reading considered so far, the sons of Sceva are particular examples of the itinerant Jewish exorcists working in Ephesus. But the Western text may tell a different story. It has been argued that ἐν οἷς in the Western text should be translated not ‘among whom’ but ‘at this juncture’ (cf. Lk. 12:1; Acts 26:12). The Western text would then suggest that Sceva’s sons are a separate group who attempt to imitate the Jewish exorcists ‘at this juncture’. Only in P38 does the Western text say that Sceva was a Jew. The corruption of the text may have arisen because once a later generation than Luke’s, ignorant of Jewish customs, failed to realize that Sceva could not possibly be a Jewish High Priest, it was almost inevitable that ἐν οἷς would be taken to mean ‘among whom’.

For the present purpose it matters less than one might at first suppose whether one accepts the Western or the majority reading here. Whether Sceva and his sons were Jews or Gentiles, Acts 19:13 has already told us what some itinerant Jewish exorcists were doing. The incident with the sons of Sceva illustrates this further whether these sons were among the Jewish exorcists or merely imitators of them.

This does not settle the question of what historical incident lay behind Luke’s account. Many commentators appear to suppose that something did, although Gerd Lüdemann dismisses it with the statement, ‘For form-critical reasons alone the joke told in vv. 13–16 is unhistorical.’ It is less important to discover whether on the particular occasion Luke describes some sons of Sceva in fact perpetrated this exorcistic fiasco than to ascertain whether it represents the sort of thing that might have happened. If it is a joke, its humour may depend on reflecting the practice of itinerant Jewish exorcists. If it is a serious object lesson, it would lose much of its point if the sons of Sceva were straw men invented for the sole purpose of being lampooned. It seems prima facie likely that Luke would aim at some level of verisimilitude, even if this particular incident is a complete fiction. Pace Lüdemann, there also seems to be no compelling reason why Luke’s account should not be based on a report of an incident that actually occurred during Paul’s stay in Ephesus, though this is more than needs to be established here. On the other hand, Lüdemann’s scepticism provides a salutary warning against trying to draw too many detailed conclusions from Luke’s account. Thus Twelftree is probably incautious in concluding that

Acts 19:13–19 tells us that the exorcists were using incantations, unaided by cultic performances, to put a supernatural restriction on demons. Their source of power-authority was the name of a renowned exorcist whose aid was sought through a careful identificatory formula.

Nevertheless, Acts 19:13–19 suggests that there were such people as itinerant Jewish exorcists, that it was at least plausible to suppose they could be found at Ephesus, and that they may have been in the habit of using incantations calling on the name of some or other power-authority in a manner similar to that indicated in some of the magical papyri. Whether or not such exorcists also employed cultic performances, and whether or not they were generally successful, is not something that may be deduced from Luke’s text.

Jewish Evidence

a. Josephus

In the course of Josephus’s account of the reign of Solomon in the Jewish Antiquities, Josephus describes both how Solomon’s wisdom encompassed the knowledge of how to expel demons, and how that wisdom was put to good use in his own day by a certain Eleazar:

And God granted him [sc. Solomon] knowledge of the art used against demons for the benefit and healing of men. He also composed incantations by which illnesses are relieved, and left behind forms of exorcisms with which those possessed by demons drive them out, never to return. And this kind of cure is of very great power among us to this day, for I have seen a certain Eleazar, a countryman of mine, in the presence of Vespasian, his sons, tribunes and a number of other soldiers, free men possessed by demons, and this was the manner of the cure: he put to the nose of the possessed man a ring which had under its seal one of the roots prescribed by Solomon, and then, as the man smelled it, drew out the demon through his nostrils, and, when the man at once fell down, adjured the demon never to come back into him, speaking Solomon’s name and reciting the incantations which he had composed. Then, wishing to convince the bystanders and prove to them that he had this power, Eleazar placed a cup or foot-basin full of water a little way off and commanded the demon, as it went out of the man, to overturn it and make known to the spectators that he had left the man. And when this was done, the understanding and wisdom of Solomon were clearly revealed (Ant. 8.45–49).

This procedure looks more magical than miraculous (at no point does Josephus speak of the glorification of God in Eleazar’s activities, although he does go on to speak of how God favoured Solomon). It is, moreover, magical in two respects. First, Solomon is represented as mastering the technique (τέχνην, Ant. 8.45) for dealing with demons. Even if God granted him the knowledge of this technique, as Josephus presents it the technique looks more like a purely human achievement (Solomon composed incantations and left behind forms of exorcisms). Moreover, τέχνη is the word used of the Egyptian magicians’ purely human art in Ant. 2.286. Secondly, Eleazar’s technique resembles the magical control of spirits. It is Solomon’s (not God’s!) name that Eleazar employs to coerce the demon, and once Eleazar has spoken the appropriate incantation, the demon is subject to his command—it can be ordered to overturn a basin of water.

Josephus appears to be trying to boast about several things at once here. First, he is trying to show what a great king Solomon was. The emphasis is not on how Solomon glorified God but on how God glorified Solomon. Secondly, Josephus wants Solomon’s glory to reflect on the Jews of his own time, not simply through being able to claim so distinguished a figure in their history, but through the lasting effects of his achievements. Thus Eleazar acts as a living demonstration of ongoing Jewish ‘wisdom’, witnessed by Josephus himself.

Given that Josephus was showing off here, one must be cautious about using this passage as evidence that Jewish exorcists were common. That Solomon’s wisdom continued down to Josephus’s own day through Jewish ability to exorcize is precisely the point Josephus is trying to establish through this anecdote. If it were already widely known that Jews made excellent exorcists, Josephus would not need to argue the point; instead, he could appeal to it as a well-known fact. But the only evidence he adduces for his statement that ‘this kind of cure is of very great power among us to this day’ is his eye-witness account of Eleazar’s performance (if this is what he is claiming, though ἱ στόρησα, which LCL translates ‘I have seen’, could mean no more than ‘I have learned about’).29 Claim and report are connected by γάρ, which suggests that the claim is being made to rest on the report. The fact that Josephus relies on his own eye-witness testimony here (effectively supported by that of Vespasian and other prominent Romans) suggests that he may be reporting something that might otherwise be viewed with scepticism, not something that was widely acknowledged. One might argue that Josephus’s point was that, since he had himself witnessed a fellow Jew casting out a demon, the procedure was not that uncommon. Yet Josephus apparently regards this particular incident as noteworthy, and makes no mention of other contemporary Jewish exorcists either here or elsewhere in his writings. Josephus’s argument is not ‘Solomon’s wisdom lives on among us because, as is well known, dozens of us are able to cast out demons using his incantations’, but ‘Solomon’s wisdom lives on among us because I know of one fellow Jew who cast out a demon using Solomon’s incantations.’ This indicates that Jewish exorcists existed; but it does not show that they were common. If they were common, one might have expected Josephus to allude to additional examples of the continuing effectiveness of Solomon’s wisdom.

The other relevant story in Josephus is his retelling of 1 Sam. 16:14–23, in which David plays the lyre to relieve Saul from the attacks of evil spirits (Ant. 6.166–69). Josephus’s most significant departure from the biblical account here is the removal of the notice that the evil spirit was sent by Yahweh. Instead, ‘strange disorders and evil spirits’ take advantage of God’s removing his divine spirit from Saul to attack him. The symptoms of these attacks are ‘suffocation and strangling’, and the effect of David’s music-making is to ‘charm away’ the spirits; although, as in the biblical account, the cure was not permanent, since David has to repeat this performance whenever the spirits attack.

In the Gospel stories, evil spirits invade and occupy the sufferer’s centre of personality. The job of the exorcist is then to drive the demon out and restore the sufferer to self-control. Josephus’s account of David’s exorcistic prowess is more akin to Pseudo-Philo’s account of the same incident (LAB 60), or the Genesis Apocryphon’s account of Abraham and Pharaoh, or the account in Tobit of the demon afflicting Sarah. Josephus does not represent Saul as losing control of his senses (unless ποιῶν ἑαυτοῦ γίνεσθαι τὸν Σαοῦλον be taken as meaning that Saul was restored to his senses, as opposed to regaining control of his respiratory functions). Instead, he suffers something like an acute asthma attack, or perhaps the kind of strangling the demon Asmodeus inflicted on Sarah’s bridegrooms (though not so terminal). Once again, the picture is one of hostile spirits inflicting suffering from without, rather than invading the psyche and causing erratic behaviour from within.

At Ant. 6.166 David is said to prophesy once the divine spirit had been transferred to him from Saul, but when it comes to dealing with Saul’s evil spirits David’s role is to ‘charm away (ἐξᾴδειν) spirits’ and play on the harp. No explicit role is ascribed to God in this charming and playing. God likewise has no explicit role in the parallel account of LAB 60. According to LAB David’s song identifies the evil spirit generically and appeals to the new womb from which he was born and to a descendant (Solomon?) as the power by which the demon is rebuked (LAB 60.3).

Josephus’s only other reference to exorcism is at War 7.185, where, in the course of describing the properties of an extraordinary root obtainable in the vicinity of Baaras, he notes

It possesses one virtue for which it is prized; for the so-called demons—in other words, the spirits of wicked men which enter the living and kill them unless aid is forthcoming—are promptly expelled by this root, if merely applied to the patients.

It is a reasonable speculation that this is the root employed in the seal of Eleazar’s ring. But what is particularly interesting here is Josephus’s view that these evil spirits are only ‘so-called’ (καλούμενα) ‘demons’—in reality they are the spirits of (presumably deceased) wicked men. This bears some resemblance to Philo’s view (that the only evil spirits were those that inhabited human bodies at some stage). It is not clear how these ‘so-called demons’ are to be related, on the one hand, to the Satanic army under the command of a demonic prince that appears to be envisaged by the Gospels (see Mk 3:20–30), and on the other, to the spirits of the disembodied antediluvian giants envisaged in 1 Enoch 6–16 and elsewhere.

Although arguments from silence are always perilous, it may be significant that Josephus mentions neither demon-possession nor exorcism anywhere else. The Gospels contain clear examples of accusations of sorcery or demon-possession levelled against Jesus by his opponents, but this is one polemical tactic that Josephus never employs. The people of whom Josephus disapproves are presented as wicked and perverse, but not as tools of the devil. The sign prophets may be false prophets, charlatans, and even sorcerers, but Josephus never suggests that they or their followers are (literally) demoniacs, although one might have supposed that this would have been a good way of discrediting them. This may suggest that although Josephus knew about both demon-possession and exorcism, he did not regard either phenomenon as being particularly common or significant. It is also interesting that in the few instances where he does talk about exorcism, the method always appears more magical than miraculous; nowhere does Josephus indicate that the power of God is behind the casting out of evil spirits, despite the fact that he clearly regards miracles as acts of God.

b. Qumran

The evidence for broadly exorcistic and apotropaic practices in the Qumran literature has already been discussed in some detail (see Chapter 7 above). Here the question to be briefly considered is whether this literature indicates anything about actual Jewish exorcists.

One candidate may be Abraham as described in the Genesis Apocryphon. It will be recalled that God afflicted Pharaoh with a plague to prevent him from defiling Sarah. When Pharaoh learns that Sarah is not Abraham’s sister but his wife, he summons Abraham to pray for him that the evil spirit might depart. Then, in the Vermes translation,

So I prayed [for him] … and I laid my hands on his head; and the scourge departed from him and the evil [spirit] was expelled [from him], and he lived.

The word ‘expel’ here certainly suggests that a demon is being driven out, though in Joseph A. Fitzmyer’s translation of the passage the same word, גער, is translated ‘rebuke’. Twelftree is well aware of the possible ambiguities of this term, and he argues that ‘One of the possible translations of גער would be to “expel”.’ He thus concludes that ‘rebuking in order to expel’ represents ‘the best understanding of what the Qumran people thought was happening in exorcism’.

We have already seen why the passage should not be read in this way. Here it will suffice to recall three points: (a) the evil spirit in this story was sent by God in response to Abraham’s prayer for assistance; from Abraham’s perspective, therefore, it is a friendly rather than an evil spirit, so this is not a story that ‘relates an individual’s ability to control and expel demons in the way we find in the New Testament’; (b) although Pharaoh is afflicted with illness, he retains full control of his faculties throughout the story; this is not an account of the casting out of an invasive demon that has taken over the sufferer’s centre of personality; (c) although the text of the Genesis Apocryphon was discovered in a Qumran cave, this does not show that the original was composed at Qumran, and since it lacks any clear sectarian features, there is little to suggest that it portrays a particularly Qumranic view of exorcism. Twelftree’s confident assertion that ‘The Qumran people understood exorcism as expelling an evil spirit’ is thus totally unwarranted. Nonetheless, the procedures attributed to Abraham here may well be based on that employed by actual Jewish practitioners, although this raises the problem of whether these practitioners were also thought capable of bringing about demonic affliction, and, if so, whose view of ‘exorcists’ this portrayal of Abraham represents.36

None of the other Qumran material portrays an exorcist at work (assuming my interpretation of the Prayer of Nabonidus is correct), although there are, as we have seen, a number of apotropaic texts concerned with defence against evil spirits. The question then is whether any of these texts presuppose the existence of an ‘exorcist’ in the sense of an individual with special skills or powers. On the face of it the strongest candidate might be 4Q510 and 511, since these are designated ‘Songs of the Maskil’. But although these texts might be intended for recitation by this particular individual, my analysis suggested that their function was more to provide a kind of liturgical screen to protect the community from demonic interference rather than to drive out any demons that had already attacked. Moreover, not only are these texts concerned with prevention rather than cure, they may also be concerned with protection against demons who lead astray rather than demons that cause illness and possession. There is thus little to suggest that the Maskil acts as an exorcist.

Scrolls 11Q5 and 11Q11 are another pair of texts that appear to indicate some kind of exorcistic activity, but they are so fragmentary that is very hard to discern precisely who is envisaged as doing what. It may well be that the ‘Davidic’ psalms of 11Q11 are being recommended for use by ordinary people as protection against demonic attacks, in which case no exorcistic specialist is in view. It is also conceivable that 11Q11 contains the text of incantations similar in conception to the Solomonic ones Josephus describes Eleazar using, in which case they might form part of the equipment of an exorcist, but this remains unclear. It is also possible that, within the Qumran context, these texts would have been seen as protection more against demonic temptations and deceptions than against possession and related forms of attack, although this is again unclear. One might also argue that the line between possession in the narrow sense and an attack on a victim’s intellect could be quite a thin one.

The amulet text preserved in 4Q560 certainly indicates a concern to counter demonic attacks, but as we saw, the intention once again seems to be prevention rather than cure. Again, an amulet is presumably intended as an apotropaic device that can be worn by anyone who feels the need. The text thus provides no direct evidence of anyone acting as an exorcist in the sense intended in this chapter. Although the existence of 4Q560 might well indicate the existence of ‘magical specialists’ who dispense such amulet texts to their clients, to label such a ‘magical specialist’ an ‘exorcist’ is to stretch the latter term too far.

Taken together, the material from Qumran provides little direct evidence for the activities of Jewish exorcists (in the strict sense) roughly contemporary to Jesus, but it does provide fairly strong evidence for a worldview in which an exorcist would have clear role to play.

c. Rabbinic Evidence

Twelftree also mentions two rabbinic references to exorcism. The first occurs in a story told in the Babylonian Talmud:

Also from the Jewish material, mention can be made of a story about the fourth generation tannaitic rabbi Simeon ben Yose. A demon, Ben Temalion, is said to enter the Emperor’s daughter. When Rabbi Simeon arrived he called out to the demon ‘Ben Temalion, get out! Ben Temalion, get out!’ The story says that as he said this the demon left the girl (b. Me ‘il. 17b.) The success of this exorcism is thought to depend entirely on the charismatic force of the exorcist.

What Twelftree neglects to point out is that in this story, the demon Ben Temalion is collaborating with Simeon ben Yose to help the rabbi secure concessions from the Emperor. Far from being dependent ‘entirely on the charismatic force of the exorcist’, the success of this exorcism relies entirely on the fact that rabbi and demon are in cahoots. Presumably this story of a demon offering to help the Jewish cause out of the goodness of its demonic heart is told tongue in cheek, which makes Twelftree’s attempt to make deductions about exorcistic technique from it highly questionable. Told out of context, Twelftree’s extract makes R. Simeon appear to be an exorcist who can, just like Jesus, expel a demon by a word of command. Placed back in context, the rabbi and the demon appear as co-conspirators in an exorcism that was a put-up job all along. The story is nevertheless of some interest, since it presupposes both the phenomenon of demon-possession and the ability of exorcists to combat it (otherwise Ben Temalion’s charade would have no point). The story is, however, too late (both in its Tannaitic setting and in its Amoraic recension) to afford any direct evidence about possession and exorcism around the time of Jesus.

Twelftree’s second rabbinic text is at least set in the first century, even though it appears in the late Numbers Rabbah:

An idolater asked R. Joḥanan b. Zakkai: ‘These rites that you perform look like a kind of witchcraft. You bring a heifer, burn it, pound it, and take its ashes. If one of you is defiled by a dead body you sprinkle upon him two or three drops and you say to him: “Thou art clean!” ‘ R. Joḥanan asked him: ‘Has the demon of madness ever possessed you?’ ‘No’, he replied. ‘Have you ever seen a man possessed by this demon of madness?’ ‘Yes’, said he. ‘And what do you do in such a case?’ ‘We bring roots’, he replied, ‘and make them smoke under him, then we sprinkle water upon the demon and it flees’. Said R. Joḥanan to him: ‘Let your ears hear what you utter with your mouth! Precisely so is this spirit a spirit of uncleanness; as it is written, And also I will cause the prophets and the unclean spirit to pass out of the land (Zech. XIII, 2). Water of purification is sprinkled upon the unclean and the spirit flees’ (Num. R. 19.8).

Once the idolater departs, however, Joḥanan privately admits to his disciples that this explanation is totally spurious: the only reason for the red heifer ritual is that God has commanded it in the torah.

The reference to roots and smoke recalls Josephus’s account of Eleazar and his description of the Baaras root. It is also reminiscent of the smoked fish gall used to drive away the demon Asmodeus in Tobit. Furthermore, the concept of possession and exorcism presupposed in the exchange is close to that of the Gospels, since the phenomenon is described in terms of ‘the demon of madness’. Moreover, Joḥanan apparently assumes that the phenomenon of possession is common enough for his interlocutor to be acquainted with it at first hand. On the other hand, the context indicates that the exorcistic technique described is a Gentile one, since it occurs in the answer that R. Joḥanan elicits from the idolater.

If one could be sure that this tradition genuinely went back to R. Joḥanan b. Zakkai, it would be a valuable witness to belief in demon possession and exorcism in the first century. Since, however, it only appears as late as Numbers Rabbah, one cannot be certain that it preserves a first-century view, even if one could be confident that there was a core of genuine Joḥanan tradition behind the story.

Although the evidence may be of dubious value, there is nothing in these stories to show that either Simeon ben Yose or Joḥanan ben Zakkai supposed casting out demons to require divine power. Since Joḥanan gives what he regards as a spurious explanation, his own views on exorcisms are not clear. Simeon secures demonic co-operation, which makes his performance effectively magical.

Later Non-Jewish Evidence

Although it perhaps lies beyond the scope of the present enquiry, we may briefly review some of the later evidence for Jewish exorcistic practice in non-Jewish sources.

Justin suggests that though Jews could in principle exorcise demons by the power of God, in practice they tended to use magical means akin to those of the pagans (Dialogue with Trypho 85). This might be dismissed as simply a piece of Christian anti-Jewish polemic if it did not agree with the impression created by Josephus’s account of Eleazar’s exorcistic practice. Irenaeus, on the other hand, creates the contrary impression, namely that Jewish exorcists operate by calling on the power of God (Against Heresies 2.6.2). Origen’s Celsus compares Jesus’ miracles to the tricks of magicians and the feats of those taught by Egyptians, who amongst other things expel demons from men in the market place for a small fee (Contra Celsum 1.68). On the one hand this indicates that Celsus regarded exorcisms as both magical (a venerated art that can be taught) and reasonably common (something that can be purchased for a few obols in the market place). On the other, Celsus associates this art not with Jews but with Egyptians (cf. CC 1.24 where Origen refers to names effective against different types of demons in either the Egyptian or the Persian tongue). Origen elsewhere (CC 5.45) remarks that ‘demons are vanquished and become submissive to him who pronounces’ the names ‘the God of Abraham’, ‘the God of Isaac’ and ‘the God of Jacob’. Presumably these names could be used by either Jews or Christians, but the lack of any specific mention of Jesus Christ may indicate a Jewish origin of the practice. Finally one should mention Lucian’s well-known ‘Syrian from Palestine’ (Philopseudes 16), conceivably a Jew, who performs exorcisms which bear some resemblance to those narrated in the Gospels (e.g. the epileptic boy and the Gerasene demoniac). This Syrian is said to adjure, threaten and charm out (ἐξᾴδοντες, cf. Ant. 6.166) the spirits, but not specifically to call on the power of any divine being. Lucian clearly intends this account to be taken with a large pinch of salt, so even if one could be sure that his ‘Syrian’ was a Jew, his evidence would need to be treated with caution. His satire nevertheless presupposes that such exorcisms are common enough to be talked about but rare enough to be regarded as spectacular. Cumulatively this evidence suggests that Jewish exorcists may have been more talked about by the second century, with differences of opinion over whether their feats were magical or miraculous. Against this it should be noted that, apart from the literature reviewed here, no surviving works of Greek or Roman authors from around the period in question mention Jewish exorcists, and only a very few allude to Jewish magic.


That many Jews (and others) in the first century ce believed in the existence of hostile spiritual forces, their manipulation by magical means, and various ways of warding them off by amulets, spells, and incantations is not at issue. What is at issue is the particular conception of demon possession and exorcism presupposed in the Gospel accounts of Jesus’ ministry. We have been able to find some indication of such beliefs outside the New Testament, just as the New Testament has itself provided some evidence for the existence of other Jewish exorcists at around the time of Jesus. Nonetheless, the evidence is remarkably slight. If demon possession and exorcisms were widespread phenomena, this is surprising.

One could argue that belief in both possession and exorcism would have been far more widespread among, say, peasants in rural Galilee than among the urban elite who were responsible for most of the texts that have survived. On the other hand, the presence of some possession/exorcism stories in writers such as Josephus, Luke and the rabbis indicates that these beliefs cannot have been the sole preserve of peasants. Moreover, given the comparative wealth of material on angels and demons in general, it is significant that there is a comparative paucity of material specifically on demon possession as the invasion by a spirit of an individual’s psyche, and on exorcism as the reversal of this condition. This hardly demonstrates that the phenomena of demonism and exorcism were especially widespread.

Such evidence as there is suffices to show that Jesus was not unique in his cultural milieu in casting out evil spirits. He may nevertheless have been remarkably unusual, not least in his claim to be doing so by the spirit of God. The impression sometimes given in modern scholarly literature that Jewish exorcists were two a penny is apparently not supported by the literary evidence. This is not quite the full story, however, and the next chapter will return to the issue of possession and exorcism in the course of exploring some anthropological perspectives.

Chapter 13

Healers, Magicians and Spirits


Although one has little option but to rely heavily on the available literary evidence, it may give a skewed view by concentrating on the surviving documents, which tend to come from the urban elite rather than the rural peasantry. This chapter will attempt to round the picture out by exploring some insights from social anthropology. The social sciences cannot be used to fill in the lacunae in the historical data. They may, however, suggest some fresh perspectives, new questions, and alternative possibilities, even if, in the end, these have to be checked against the same literature.

Not all the literature from this period is the product of urban elites, however, and this is especially so of the New Testament. Since this study is concerned to illuminate the miracles of Jesus, it is this literature that will be particularly in view as we investigate the contributions of medical anthropology and of social-scientific research into witchcraft and spirit possession, which together form the three main topics of this chapter. It will not be possible to develop these themes in full here; the aim is rather to provide a sketch of how the data might appear from a perspective different from those employed hitherto.

Healers (Medical Anthropology)

Medical Anthropology is that branch of social anthropology that studies societies from the point of view of health care. Its application to the study of Jesus’ healing activity has been pioneered in particular by John Pilch. This chapter will not engage directly with Pilch’s work, but will rather focus on three areas to which it calls attention. These are the aetiology of sickness, the distinction between illness and disease, and the model of the Health Care System.

a. Sickness Aetiology

One way of classifying beliefs about sickness is to enquire whether a particular system is personalistic or naturalistic, that is whether the people who subscribe to the system blame sickness on personal agents or impersonal, ‘natural’ causes. George Foster proposes that ‘disease aetiology is the key to cross-cultural comparison of non-Western medical systems’. By this he means that many other features of these systems can be correlated with beliefs concerning the cause of sickness.4

One might suppose that first-century ce Jews, in common with many other persons in the first-century Mediterranean, held a mainly personalistic view in which sickness might be due to either divine punishment or demonic attack, or perhaps to sorcery or the evil eye. Curiously, however, many features of Jesus’ reported healings fit the naturalistic side of Foster’s taxonomy better than the personalistic. According to Foster, in personalistic systems a wide range of misfortunes, including but not restricted to sickness, are explained in the same way, so that the healer will typically be concerned not only with curing sickness but, for example, with helping people to cope with blighted crops, financial reverses or soured marital relationships. Apart from the occasional rescue and feeding, however, Jesus is never presented as addressing personal problems other than sickness. When approached with other types of problem, such as the division of an inheritance (Lk. 12:13–15), Jesus refused to address it on the petitioner’s terms. Jesus’ miracles of healing, by being restricted to cases of sickness, conform to a naturalistic system. They also do so by performing a therapeutic rather than a diagnostic function. In a personalistic system the primary role of the practitioner is to identify the agent responsible for the sufferer’s condition; this Jesus never does. In common with practitioners in naturalistic systems, Jesus concentrates on healing the sickness. On the other hand, when it comes to preventive measures, Jesus never teaches people to avoid potential health hazards such as cold floors and chill breezes, but rather urges them to ensure that their social networks with fellow human beings and with God are in good order. This places him on the personalistic side of Foster’s taxonomy, unless one argues that Jesus’ teaching is not directly linked with his healing activity in this way.

It may be that Foster’s taxonomy simply breaks down in first-century ce Palestine. But it is not inconceivable that Jesus should have operated within a ‘naturalistic’ system in Foster’s terms. For one thing, a naturalistic system is perfectly compatible with official belief in a Deity who is ultimately held to be responsible for all that happens, or who may be petitioned for aid against sickness.9 For another, by Jesus’ time Palestine had been subject to Hellenistic influence long enough for it to be at least conceivable that a popularized form of the ‘great tradition’ of Greek medicine had filtered down. Finally, belief in demons does not automatically make a system ‘personalistic’ in Foster’s sense, especially if the demons are conceived of both as part of the normal furniture of the universe and as attacking randomly. To be seen as part of a personalistic system the demons would have to be supposed either to have clear motives for singling out particular victims, or to be under the control of some agent who wished to attack particular persons. Purely capricious demons are effectively as naturalistic as randomly striking microbes.

In a naturalistic system, religion and magic play a limited role in the explanation of the cause of disease. On the whole, Jesus seems to have been reluctant to blame disease either on God or on human agents. This raises the question of what disease aetiology is presupposed in Jesus’ healing activity. In what follows we shall find little evidence that either sorcery or evil eye belief played a major role. But there is also little to suggest that Jesus concerned himself with any ‘natural’ causes of sickness. He is never represented as healing such complaints as cuts or broken bones, where the natural cause (accident or human aggression) would be most apparent. The role accorded to physicians and medicines by both Philo and Sirach may suggest some recognition of natural cause of disease, at least among educated Jews, even when both authors insist that God plays a vital role in healing. Elsewhere we have seen illness attributed either to demonic aggression or divine punishment. Since Jesus appears to have repudiated the latter explanation (Jn 9:3; cf. Lk. 13:1–5), this may indicate that he attributed all illness to the former; but this is a question to which we shall return below.

b. Disease and Illness

The medical anthropologist Arthur Kleinman observes a careful distinction between ‘sickness’, ‘disease’ and ‘illness’. For Kleinman, ‘sickness’ is the generic term covering the other two. ‘Disease’ is the biomedical (or psychological) malfunction afflicting an individual organism. ‘Illness’, on the other hand, is the sociocultural construction placed on a sickness by the sufferer’s own ‘explanatory model’ of their condition. Whereas disease affects only the afflicted individuals, illness may affect many other people besides. It may, for example, directly involve the sufferer’s friends, family, and wider community. Thus, whereas disease is considered an objective, value-free condition that can be determined purely scientifically, illness is at least partially subjective, and may well carry certain cultural values.

Corresponding to this distinction between disease and illness, Kleinman makes a further one between curing and healing. A disease is cured when a therapist intervenes biomedically to correct the physical (or psychological) malfunction. An illness is healed when the sufferer (and those connected with him or her) is satisfied that the problem has been dealt with, a process that involves identifying the illness, giving it a socially meaningful identity and explanation, and applying a culturally appropriate remedy.

Kleinman makes it clear that his distinction is an etic one devised for the purpose of cross-cultural comparison. The cross-cultural comparison he is particularly interested in is that between the modern West and modern Taiwan, but many of his observations, and particularly his model, may be equally applicable to the ancient Mediterranean.

One question that particularly exercises him is why indigenous folk-healers such as the Taiwanese tâng-ki (or shaman) often prove more effective at healing illnesses than Western-style doctors operating in the same culture. A major part of his explanation is that many illnesses—even those that appear to be bodily ones—in fact have mental causes. According to Kleinman, in Taiwan and the USA many patients—up to 50 per cent—’suffer from somatic complaints that are due to psychological problems’. Moreover, in his opinion, most illnesses are in fact anxieties about minimal symptoms, not due to serious disease but to psychosocial stressors, which may encourage the sufferer to enter into the socially-sanctioned sick role. This leads him to state that ‘Therapeutic response in these cases will obviously follow removal of the psychosocial stressor and reduction of anxiety.’15

In Taiwanese culture considerable stigma attaches to any form of mental illness. People with mental or emotional difficulties are reluctant to admit to them, with the result that they tend to ‘somatize’ their problems (that is, experience them as some form of bodily disorder), since physical illness is far more socially acceptable. Traditional healers such as the tâng-ki are skilled at dealing with this kind of problem through rituals that effectively provide a form of indigenous psychotherapy. In particular, these indigenous healers take their clients seriously by providing explanations for their suffering in terms that they can relate to, unlike the Western-style doctors who take what Kleinman castigates as a ‘veterinary’ approach.

Kleinman believes that these conclusions are by no means confined to Taiwan, and that somatization of psychological problems is also prevalent among poorly educated people in the West. It therefore seems at least possible, if not probable, that the same would have been so for Galilean peasants in the first century. Again, a high proportion of the illnesses Jesus is reported as healing (blindness, deafness, dumbness, fever, various forms of paralysis, and possession) could be due to somatization or conversion hysteria. It is at least plausible, therefore, to understand Jesus acting as a healer of psychosomatic illnesses in the manner of a first-century Jewish equivalent of a Taiwanese tâng-ki. Jesus often achieved more than the cure of a disease. In such cases as the cleansing of a leper or the curing of the woman with a haemorrhage, the cure would also have permitted the sufferer’s integration into society, so that social relations will have been healed as well as a disease cured. In four stories (Jairus’s daughter, epileptic boy, Syrophoenician woman’s daughter and widow of Nain’s son) children are restored to their parents. The cured paralytic returns home, the Gerasene demoniac finds a new missionary role, and blind Bartimaeus follows Jesus. Peter’s mother-in-law is restored to her domestic role. In all these stories one glimpses a healing along with the cure. Jesus’ reply to John the Baptist (Mt. 11:5) includes the preaching of good news to the poor alongside the cures of various diseases, which places these cures in a wider, social, healing context.

On the other hand, the Gospel accounts of Jesus make him appear very unlike one of Kleinman’s Taiwanese shamans. Quite apart from the lack of ritual, there is no account of Jesus inquiring into the wider social, emotional or family situation of his patients. Each reported healing is far too rapid for any effective psychotherapy—including folk psychotherapy—to have taken place. Again, Jesus can hardly have been healing psychosomatic illnesses in cases where he revived people thought to be dead (even if they were only unconscious). Finally, the miracle stories focus on the cure of the disease, not the healing of the illness: blind people see, deaf people hear, lame people walk, but we learn little about other aspects of their life situation.

c. The Health Care System

In his discussion of Taiwanese medicine, Kleinman also introduces the model of the ‘health care system’. By this he means an etic construct for the purposes of cultural comparison, not an entity such as the National Health Service. A ‘health care system’ in Kleinman’s terminology is simply a society analysed from the perspective of health care.

Kleinman’s model divides all health care systems into three sectors: professional, popular and folk. The professional sector comprises all those who practise medicine with official accreditation. How far such a definition is applicable to the first-century Roman Empire is questionable; although there were people who made their living as physicians, it is not clear what would count as official accreditation; perhaps one might envisage persons who had undergone training in the tradition of ‘scientific’ Graeco-Roman medicine. The popular sector comprises the non-specialist care provided by friends, relatives, neighbours and, indeed, the sufferers themselves. In this sector illness is treated according to the remedies generally known and widely available in the particular culture (in our culture, for example, staying in bed and dosing oneself with aspirin). Kleinman estimates that between 70 and 90 per cent of all illness episodes in Taiwan and the USA are handled in the popular sector. It is unlikely that this proportion would have been less in first-century ce Palestine.

The folk sector, which overlaps with the other two, comprises persons who specialize in healing to a greater extent than the common remedies supplied by friends and relatives but who are not officially recognized as professional physicians. In a society that lacks the political apparatus to accredit professional physicians, any specialist healer (e.g. shaman, medicine man, or curandero) will be a folk-healer, but folk-healers commonly exist in societies that possess a professional sector as well. In respect of his healing activities, Jesus plainly belongs in the folk sector.

This classification does not take one very far unless one can say anything about folk-healers in first-century Palestine. Presumably they existed, since they tend to exist in all societies, and there would have been as great a need for them in Palestine as anywhere else. It may be that the πολλῶν ἰατρῶν of Mk 5:26 refers to folk-healers rather than professional physicians, since the woman probably did not come from the stratum of society that had ready access to the latter. The strange exorcist of Mk 9:38 may also have been a folk-healer of sorts, as, conceivably, may Hanina ben Dosa. The evidence for healing by laying on of hands we reviewed in Chapter 7 may be another indication of folk-healing activity. In 1 and 2 Kings both Elijah and Elisha are portrayed as fulfilling folk-healing roles on occasion, which suggests that it was at least possible for the role of folk-healer to be combined with that of popular prophet. Unfortunately, this evidence is far too meagre to enable one to sketch out a portrait of the typical Galilean folk-healer. Indeed, these figures all appear to be very different from one another, as they are from the Taiwanese folk-healers described by Kleinman.

There is, however, one line of questioning that may prove fruitful. Given that there probably were other folk-healers operating in Galilee at the time, what made Jesus stand out as exceptional? A possible answer lies in the model Octavio Romano developed in relation to Latin-American folk-healers (curanderos) and applied especially to the Mexican folk saint Don Pedrito Jaramillo, who ministered in South Texas from 1881 until his death in 1907. There were many other folk-healers operating in the same area, but none achieved anything like the reputation of Pedrito Jaramillo, whose shrine was still being visited by a steady stream of pilgrims fifty years after his death.

One must be cautious in employing Pedrito as a parallel to Jesus’ folk-healing role, since Pedrito, a devout Catholic, may well have modelled himself on Jesus. Nevertheless, the model Romano proposes to explain how the curandero became a folk-saint does contain some suggestive parallels with Jesus’ ministry. Moreover, there are sufficient similarities between the Mexican peasant culture described by Romano, and the Mediterranean culture described by, say, David Gilmore to make a comparison worth while. This is not just a coincidence; the Mexican peasants Romano describes are Spanish-speaking Roman Catholics who have been heavily influenced by Mediterranean culture.

One universal feature of Mediterranean culture that appears to have been reproduced in nineteenth-century southern Texas is the tendency towards social atomism. Romano describes this as ‘tendencies toward individual independence, autonomy, and uncooperativeness’. He proceeds to spell out what this means for male Mexican peasants in slightly more detail:

Briefly, the adult male should be wary of an undependable social environment, and he should sustain his personal autonomy and independence by generally withdrawing from, and avoiding, close associations with non-kin. If the male is successful in following this pattern of non-interference and avoidance, while also sustaining a position as head of an extended family, he may gain the respect of the community’s citizens. This respect is symbolized by their granting him the respect title Dom … The role definition of male autonomy, therefore, tends toward an atomistic social order which emphasizes mistrust and suspicion as well as general uncooperativeness.

This description does not appear to be wholly unlike that given of the agonistic society of the Circum-Mediterranean, just as the notion of ‘respect’ that features in this description appears to be akin to the Mediterranean concept of ‘honour’, even if it is not totally identical to it.

This is only one side of the picture, however. Working against the tendency towards social atomism are the behavioural requirements for mutual assistance, not least ‘from one’s neighbours, from a patrón, from proprietors of stores …’ A certain degree of co-operation is necessary if people are going to live together in society. Amongst Catholic Mexican peasants there also seems to have been an awareness that the Christian ideal called for a degree of selflessness. If one substitutes ‘Israelite’ for Christian in the previous sentence, one could well imagine similar contradictory tendencies at work amongst peasants in first-century Galilee. The interplay of these conflicting tendencies is a crucial part of Romano’s model.

Romano begins by locating the healer role within the context of mutual assistance. If a (male) Mexican peasant adopts the role of healer, he is forced to orient himself towards the co-operative pole, since the healer is expected to be at the service of the public. Someone who intends to persist in the role of healer is obliged to become increasingly oriented towards communality. Moreover, the further up the ‘healing hierarchy’ he moves from family, through village, district and internationally renowned healer, the more he must renounce himself and dedicate himself solely to the role. For a regional healer, ‘Renunciation of multiple roles by the healer, along with alternative goals, now begins to become a way of life.’ Such ‘multiple roles’ would include family life and earning one’s living through normal economic activity. Such a renunciation is not unlike that practised by Jesus himself and demanded of his immediate followers. It also characterizes Pedrito Jaramillo.

The question remains what made Pedrito stand out from other curanderos. Romano suggests a number of factors. First, unlike other curanderos, Pedrito claimed a direct divine license for his healing activity, without saintly intermediary. Secondly, his fame was probably spread by his extensive travels round Texas in the performance of his role and by his providing food for those who came to see him. Thirdly, his healing methods were strikingly individual, usually involving instant prescription of the first thing that came into his head, which might vary from coffee to tinned tomatoes but was most usually water. All these factors were taken up into the cycle of healing stories that began to circulate even in Pedrito’s lifetime.

Yet, in Romano’s opinion, none of these factors fully explains why Pedrito exercised a charisma that other curanderos lacked, since many of them were also innovative in their methods. What made Pedrito stand out in particular was not his innovations, but his ‘performing his role in relatively strict accordance with the fundamental and generic definition of healer role as provided by tradition, which relates to the behavioral sector of communality and mutual assistance’. Pedrito put into practice a dormant ideal of the totally selfless healer, and thereby ‘singularly reasserted tradition by making the pre-existent “ideal” into a tangible and recognizable entity’. He is thus to be regarded as a charismatic renovator embodying the charisma of conservatism. As such, however, he might just as easily have met resistance as a revolutionary innovator would, since the attempt to reintroduce a pre-defined form can be as much as challenge to the status quo as the attempt at radical innovation.

With this should be compared an article by Bruce Malina that questions whether Jesus should be called a charismatic leader in the Weberian sense. Malina argues that in Weberian terms Jesus was not a charismatic (revolutionary) leader but a reputational (conservative) one, concerned with the restoration of Israel and its traditional values. His elevation from honourable Galilean healer into revered national Messiah was due to his satisfying some principal aspirations of Palestinian society. Jesus’ success in defending his honour, together with his healing ability, made him an effective symbol for non-elite persons who saw their own honour being trampled. Jesus repeatedly expressed indifference to personal power over others. In a first-century ce Jewish context, Malina maintains, it was this refusal to assume power than made him a moral hero, committed to the Kingdom of God.

Malina and Romero are plainly not saying precisely the same thing, but their arguments mesh nicely. Jesus’ refusal to assume power could be seen as part of a renunciation of alternative roles, just as his satisfying some principal aspirations of Palestinian society would be congruent with his realizing a dormant social ideal. The question is then what ideal he was realizing; for Jesus, unlike Pedrito, was hardly a folk-healer simpliciter. In Jesus’ case it might be nearer the mark to say that the dormant role he realized was that of the servant, or agent of God, partially exemplified in the Old Testament by such ‘men of God’ as Elijah and Elisha who combined the folk-healing role with a prophetic one. All this, together with the communal emphasis of the folk-healer role, would fit in well with a view of Jesus that saw him as prophet concerned with the restoration of Israel.


The issue most commonly discussed in relation to Jesus and magic is whether Jesus’ miracles were ‘magical’ or in what way they differed from ‘magic’. This issue will be briefly examined here along with another that arises out of the discussion of the aetiology of sickness, namely whether there is anything to suggest that Jesus was countering magic in his healings.

Whereas writers on antiquity tend to speak of ‘magic’, social anthropologists more often talk about ‘sorcery’ or ‘witchcraft’, sometimes, but not always, distinguishing between the two. This distinction stems from E.E. Evans-Pritchard’s work on the Azande. Among these people ‘witchcraft’ is the power of a special class of people (witches) to inflict harm on others by malicious wishes alone; this power may be exercised consciously or unconsciously. ‘Sorcery’, on the other hand, is a technique that in principle can be practised by anybody to achieve magical ends (though one may prefer to employ a specialist), and makes use of incantations, spells and magical materials (including ‘medicines’ or poisons). On this understanding, witchcraft is an innate ability, while sorcery is an acquired technique. It should not, however, be supposed that this distinction can automatically be universalized beyond the Azande.31 Moreover, even writers who are aware of the distinction do not always observe it, often, for example, using ‘witchcraft’ as a blanket term to cover both forms of magic. Again, even when it is possible for the observer to make a distinction along these lines, it may turn out to be a distinction of little practical importance in the society observed.

In what follows I shall use ‘witchcraft’ to refer to harmful magic performed by innate ability and ‘sorcery’ to refer to magic performed by acquired technique. Most of what is called ‘magic’ in discussions of antiquity resembles ‘sorcery’ more than ‘witchcraft’. On the other hand, the evil eye, belief in which is widely attested for the Mediterranean region in general and the biblical Jews in particular, is logically a species of witchcraft, since it is supposed to be the innate ability of certain people to harm persons or property just with a look.

Most writers on witchcraft and sorcery stress that beliefs about witches and sorcerers are widespread both in time and space, so that there is considerable continuity in magical beliefs between classical antiquity and subsequent periods, and considerable similarity between classical and mediaeval European beliefs and those held by primitive peoples in the modern world. This suggests that anthropological studies on present-day societies may yield useful insights into the beliefs that were held in the first-century circum-Mediterranean. Three insights in particular stand out from such studies. The first is that beliefs in witchcraft or sorcery tend to function as explanations for misfortunes that are otherwise inexplicable, particularly where neither the vagaries of chance nor one’s own inadequacies are acceptable explanations. The second is the function of witchcraft (or sorcery) accusations as providing socially acceptable ways of attacking one’s enemies (usually one’s equals or inferiors, seldom, it seems, one’s superiors unless one is aiming to displace them from office). In most anthropological studies, the alleged ‘sorcerer’ is regarded as the victim, while his or her alleged ‘victim’ turns out to be the real aggressor. The third is the importance of the ‘witch-finder’ or ‘cunning man’ in focusing witchcraft (or sorcery) suspicions.

In asking whether or Jesus’ miracles were magical, one must first decide whether the distinction one is trying to make is an emic or an etic one, or at least be clear at any one point which one has in mind. For the present purposes it is the emic distinction between magic and miracle that is of particular interest, since this study is primarily concerned with how Jesus’ miracles would have appeared in their Jewish context.

A number of attempts have been made to distinguish magic from miracle in this context. Some focus on alleged magical technique, some on the aims of magic, and many attempt a sociological definition. None of these approaches is quite correct if one is concerned to pin down the emic distinction.38 Josephus, for example, does not represent Moses and the Egyptian magicians as employing visibly different techniques when they turned their rods into snakes before Pharaoh. Neither is it obvious to Pharaoh that they are pursuing radically different aims. Again, attempts to define magic solely in terms of social deviance beg the question of from whose perspective something is to be considered deviant, and fails to explain why one particular form of deviance should be labelled ‘magic’ rather than something else equally pejorative.

An adequate emic definition of magic in a first-century ce Jewish context requires three dimensions, not just one, the three being sociological, phenomenological and theological. Of these the third is the probably the most important. Whereas magicians are certainly regarded as deviant, and accusing people of magic can therefore be a way of discrediting them, for the accusation to be credible, more than mere deviance must be shown. The particular form of deviance is the ability to perform extraordinary deeds beyond normal human ability; this is the phenomenological dimension. The theological dimension is that the source of power for these extraordinary deeds is believed to be other than God. This is precisely the point at issue in the Beelzebul controversy, and, as we shall see, it is precisely the point at issue in other Jewish texts.

For Josephus, the key difference between Moses, Aaron and the Egyptian magicians who replicate their signs is that Moses’ signs are performed by no human art but by the power of God (even though this is apparently not something Pharaoh is able to verify empirically). For Philo, this same incident furnished proof that God was behind Moses and Aaron (Philo thinks this can be verified from the greater effectiveness of divinely wrought miracles). Deuteronomy 18:9–14 lists a series of ‘abominable practices’ forbidden to the Israelites, which includes terms for various types of sorcery. Deuteronomy 18:15–19 then goes on to describe the prophet raised up by Yahweh as the figure who should be heeded, which suggests that the redactor wished to contrast the two: the legitimate source of superhuman knowledge is the God-inspired prophet, all else is condemned as sorcery. Deuteronomy 13:1–5 likewise prescribes the death penalty for any would-be prophet who works signs and wonders to lead people astray from Yahweh. It may well be such passages as these that lie behind the sketches of the demonically-inspired magicians and deceivers we noted in Chapter 11. It is also noteworthy that Josephus sometimes equates the false prophet with the γόης in his discussion of the sign-prophets. Although γόης can simply mean ‘charlatan’ or ‘deceiver’, it can also mean ‘magician’, and even if that was not the sense Josephus had primarily in mind, by using the word he manages to convey the impression that any sign worked by these false prophets would have been magic, not miracle.

In the Jewish context, then, the decisive distinction between magic and miracle is not something directly observable, but whether the surprising feat was held to be wrought through God’s power or through some other means (diabolical power or human technique). To judge whether or not Jesus should or should not have been called a magician, then, is to make a decision that can only be made on the presupposition that there is a God whose power can be made manifest in surprising events, and then through one’s belief whether or not Jesus was in fact an agent of that God. If one wishes to maintain (or at least feign) neutrality on either of these issues, one can only point out what is at stake in the question; one cannot then go on to answer it.

This is not to say that, having settled what was at stake in labelling someone a magician, one may then not go on to investigate the general characteristics of persons so labelled. Here discussions of technique, aims, and social dynamics may well become relevant provided that they are treated as indicators rather than defining characteristics. On this understanding, Theissen’s sociological treatment, which divides doers of ‘miraculous’ deeds into institutional (e.g. at healing shrines), magical and charismatic is both legitimate and helpful. One must bear in mind, however, that his etic sociological classification will not entirely correspond to my emic distinction. It is perfectly reasonable to point out that magicians tended to work in secret to avoid detection whereas charismatics sought publicity.44 But this is not a distinction that would have impressed the Jewish authorities as a defence of charismatics; Jesus exorcized publicly but could still be accused of doing so by the power of Beelzebul; likewise, in Josephus’s eyes the sign prophets did not legitimate themselves by promising to perform their signs in public.

It is also legitimate and helpful to criticize modern attempts to label Jesus a magician by pointing out the differences between the miracles of Jesus portrayed in the Gospels and the type of procedures envisaged in the Greek Magical Papyri and other comparative literature. But again, this is unlikely to have been a comparison that would greatly have exercised Jesus’ Jewish contemporaries. The only point of practical concern was whether Jesus was a true man of God or a diabolical counterfeit, a true prophet or a false one (cf. Mk 11:27–33). In deciding this the authorities who accused him of sorcery were probably far more concerned by the challenge to their own authority implicit in his message and his following than in the details of his thaumaturgical technique (although the motive for accusation should not, of course, be confused with the nature of the alleged offence). The fact that on some occasions Jesus may have used saliva, mud or certain verbal formulae is no more relevant to such an assessment than the fact that Tobias used the innards of a fish. In the end, Jesus’ claim to come from God could only be judged by whether his deeds and words appeared consistent with such a claim.

The reported miracle of Jesus that looks most magical is the cursing of the fig tree (Mk 11:12–14, 19–21). That Jesus performs an act of destruction by a curse certainly resembles witchcraft, and the episode is startlingly similar to withering attributed to the evil eye in modern rural Greece. Mark’s redaction neutralizes its magical overtones in three ways. First, by intercalating Jesus’ prophetic sign against the temple between the two halves of this miracle, he turns the destructive miracle into a legitimate prophetic sign. Secondly, by immediately following it with a discussion on faith and prayer he suggests that it was divine power at work even in this act of seeming thaumaturgical vandalism. This incidentally strengthens the case that the real point at issue in distinguishing magic from miracle was the source of the power. Thirdly, therefore, Mark follows the second half of the fig tree story with an explicit discussion of the source of Jesus’ ἐξουσία (Mk 11:27–33). Mark’s treatment of this episode thus suggests that he felt the need to domesticate a well-known story that made Jesus appear embarrassingly like a witch. It is possible that the story arose from a mis-understanding (or malicious misrepresentation) of some incident in Jesus’ ministry; or it could even be that Jesus in fact cursed a fig tree which withered soon after. What can be argued with greater confidence is that if the fig tree incident was originally told against Jesus to defame him as a magician, it was the worst his detractors could come up with. There is no suggestion that Jesus ever employed magic against human beings (or animals).

There is little to suggest that Jesus considered himself to be combating human magical opponents either. Never in his healing activity does he play the role of ‘witch-finder’, helping his clients to identify the enemies whose sorcery had made them ill (as one would expect in a personalistic belief system). It may be that he deliberately avoided doing so to avoid the socially divisive consequences of sorcery accusations, just as he avoided adding to sufferers’ burdens by suggesting that their afflictions were punishments for sin (the apparent exception at Mk 2:9–11 may be no more than an ad hominem argument aimed at the scribes).

It is conceivable that evil eye beliefs are involved in the concern for privacy associated with some of the miracles. If either Jesus or the sufferer supposed the illness to be due to the effects of the evil eye, then it would make sense for the healing to take place away from potentially baneful gazes. The fact that Mark makes his own use of the secrecy motif does not in itself exclude the possibility that he occasionally found it in his sources. But there is nothing that compels us to understand the secrecy motif in terms of the evil eye.

In the main, the Gospels are silent about the supposed cause of the illnesses Jesus heals. One could argue that this is the result of the smoothing out of miracle stories in the course of oral transmission, since the tradition was primarily interested in Jesus rather than in the medical history of his patients. On the other hand, although no magical opponents appear in the Jesus tradition, they do occasionally appear in Acts. Acts 8:9–24, for example, describes the apostles’ encounter with the magician Simon. The incident with the sons of Sceva concludes with a great many magical books being burned (Acts 19:13–19), and we have already seen how the Jewish exorcists appear to have been attempting magical techniques. Acts 13:6–12 recounts Paul’s conflict with a Jewish magician named Elymas. It is noteworthy that this same passage also calls this man both a ‘false prophet’ and a ‘son of the devil’. Finally, the slave-girl with the spirit of divination at Acts 16:16–19 appears to have been exploited by her owners for broadly magical ends. The silence about magicians in the Gospels may therefore be more than a mere accident of transmission-history.

The one Gospel pericope where sorcery accusations are most plainly in view is the Beelzebul controversy, which presents almost a paradigm case of authorities trying to discredit a potentially dangerous upstart by accusing him of what, in a Jewish context, would count as sorcery (performing superhuman feats through an agency opposed to God). But this account leads us into the realm of spirit possession, which forms the subject of the next section.

Spirit Possession

‘Spirit possession’ is a category that encompasses demonic possession, but includes many other things besides. Demon possession is the form of spirit possession that a particular society regards as pathological, an illness that needs to be dealt with. But spirit possession can also include phenomena that particular societies evaluate positively, including shamanism (or, more generally, spirit control) and spirit-mediumship. In connexion with the New Testament the potential interest of spirit possession extends beyond demonic possession into fields such as glossolalia, inspired prophecy, and the ability to heal and work miracles (cf. 1 Cor. 12:9–10). It also shares a frontier with witchcraft and sorcery. Most generally, spirit possession may be defined as ‘any altered state of consciousness indigenously interpreted in terms of the influence of an alien spirit’. This could also include apocalyptic visions, since the notion of an ascent to heaven to learn previously hidden secrets is not wholly unlike the spirit-journeys of classical shamanism.49

Possession states are by no means necessarily pathological. From the anthropological or psychological viewpoint they are more properly understood as the culturally constructed way in which many societies understand trance states that are in principle attainable by all human beings. The dissociative mental state that is often labelled ‘possession’ is neither a form of mental illness nor a peculiarity of primitive savages, but simply an innate capacity of the human brain. Far from regarding them as pathological, some writers see such dissociative states as positively therapeutic. They may also be understood as akin to a dramatic performance with a culturally coded script involving ways of knowing and of imaging the self that are foreign to the assumptions of Western individualism.52

One should therefore be wary of assimilating demon possession to the forms of mental illness commonly recognized in modern Western societies. It should not automatically be labelled ‘schizophrenia’ or even ‘hysterical psychosis’, Neither is it obviously the same thing as the condition labelled ‘Multiple Personality Disorder’ in Western psychiatry; MPD is normally the result of a trauma in the past (typically in childhood) whereas possession is frequently a response to stresses in the present. Again, persons suffering from MPD generally have many alternative personalities, whereas the demon-possessed normally have only one.54 To be sure, a culture will not label something ‘demon possession’ unless it regards the behaviour of the possessed person as worrying, and it may well be that mentally ill persons who become possessed will exhibit particularly worrying symptoms; their dissociative state may well allow their illness out into the open. One might thus regard the more spectacular manifestations of demon possession as a culturally coded way of expressing and dealing with the social, mental or emotional stress that is the sufferer’s real affliction. In that sense, demonic possession may be symptomatic of an underlying mental disorder, but it is not itself a pathological state simply by virtue of being dissociative. Possession behaviour may be part of the cure rather than part of the disease, even if it is seen as part of the illness.

Demonic possession may also occur when there is no underlying disease, as a response to domestic, social or political stress. The fact that possessed persons are regarded as demoniacs may then be more a cultural evaluation of their protest against oppression than an accurate assessment of the state of their mental health. For example, I.M. Lewis’s study of what he calls ‘peripheral possession’ focuses on situations of domestic stress. Here the typical sufferer is a woman downtrodden by her husband, who turns the tables by acquiring a spirit that can demand attention and gifts in its persona that would be ignored if demanded in her own. In this way peripheral possession functions as an oblique strategy of attack, rather like accusations of sorcery. The difference is that whereas sorcery accusations are generally levelled at equals or inferiors, possession is nearly always employed by inferiors against their social superiors, as a means of redressing the balance of power. In the main the superiors tolerate this strategy, perhaps because it is seen as a safety valve that does not seriously challenge the status quo. But when possessed persons accommodate themselves to their possessing spirits and graduate to becoming spirit mediums or spirit controllers (e.g. shamans or exorcists), they open themselves to charges of sorcery, since they may now be seen as effective leaders of protest against oppression. The suspicion is that someone who can control spirits is probably in league with spirits. Accusing such people of sorcery is a way of discrediting them and limiting their influence. Once again, the Beelzebul controversy comes to mind (although on Lewis’ model the spirit-controller would normally become the leader of a peripheral possession cult helping possessed persons to form a constructive relationship with their possessing spirits, whereas Jesus simply casts the spirits out).56

Such ‘peripheral’ possession is not limited to cases of domestic stress. Even in the domestic situation it may be seen as enhancing what women can contribute positively rather than as a form of protest by the marginalized. It may also be employed for purposes ranging from self-punishment through self-aggrandisement to control of other people.58 It can also occur in societies under acute stress where the whole society is in danger of becoming marginalized, or in circumstances of radical social change, or as a form of resistance to colonialism and imperialism. Since any or all of these factors could have been operative in first-century ce Palestine, it would be rash to insist that the demon-possession encountered by Jesus must have been due to any one of them. That said, the apparent absence of demon-possession in Palestinian Judaism prior to the first century, coupled with increasing social, economic and political pressure on the Jewish people, could suggest a link between the two.

Thus far this study has suggested that demon possession and exorcism were relatively new phenomena in Judaism at the time of Jesus. But this cannot be said of spirit possession in general, which appears to have been known in Israel in Old Testament times. Many of the prophets appear to have prophesied in a state of ecstasy that would be equivalent to spirit possession or spirit mediumship. Deuteronomy 18:9–14, which is concerned to contrast forbidden divinatory practices with legitimate prophecy, envisages various types of spirit-related phenomena. There are also references to such mediumistic or necromantic practices at, for example, Lev. 19:31; 20:6, 7; Isa. 8:19 and 1 Sam. 28:3. Prophecy and charismatic leadership in war are two types of spirit-filled activity that the Old Testament generally regards as legitimate, but even prophetic behaviour was not always evaluated positively. For example, at 1 Sam 16:1–13 David is anointed king in Saul’s place. The spirit of Yahweh at once departs from Saul and an evil spirit is sent in its place (1 Sam 16:14), which seems to be a way of saying that Saul’s spirit-possession turns demonic. The evil spirit causes Saul to ‘act like a prophet’ (ויתנבא) in an apparently uncontrolled and violent manner. At 1 Sam. 19:20–24 first Saul’s messengers and then Saul himself end up acting like prophets, which means that they are incapacitated and thereby prevented from reaching David. When Saul comes to Naioth in Ramah in pursuit of David and Samuel, he strips off his clothes and lies naked and helpless all day and night. This is then given as an alternative explanation for the saying, ‘Is Saul also among the prophets?’ Here typical prophetic behaviour is evaluated negatively as uncontrolled and incapacitating (contrast 1 Sam. 10:5). This suggests that Saul was thought of as acting in a spirit-possessed manner that, by the first century, would have been labelled demon possession. The difference is that the Old Testament lacks the later dualism in which the evil spirits are conceived as forming an independent and rebellious realm under a satanic leader opposed to God. Nevertheless, the notion of demon possession appears to be present in embryo. Prophets and charismatic leaders of whom the writer approves are said to have been seized by the spirit of Yahweh; those punished by uncontrolled or erratic behaviour have been sent an evil spirit by Yahweh. Zechariah 13:2–6 provides a good post-exilic example of spirit-inspired prophecy receiving a strongly negative evaluation from a writer who regarded the inspiring spirits as unclean.

It seems likely that this dichotomy between legitimate and illegitimate forms of spirit-possession continued uninterrupted into New Testament times, although it became intensified through a growing demonology, which no longer saw the ‘evil spirits’ as subject to God’s control. Evidence for some continuity in the notion of prophecy as spirit-possessed inspiration comes from Philo, who frequently suggests that in prophecy the prophet’s persona is replaced by another (Vit. Mos. 2.188, 191; Rer. Div. Her. 258–59, 264–66; Mig. 34–35; Spec. Leg. 4.49; Quaest. in Gen. 3.9). Josephus’s equation of demons with the spirits of wicked dead humans (War 7.185) may also be an echo of the Old Testament evaluation of the oboth consulted by mediums. This, together with the charismatic phenomena suggested by the Gospels, Acts and some of Paul’s letters, suggests that people continued to exhibit forms of ecstatic behaviour. In Palestine these were normally frowned upon; where the behaviour appeared pathological it was regarded as possession by an ‘unclean spirit’ or demon; in other cases, where it gave power to the possessed, it might be regarded as sorcery, an illegitimate attempt to consort with spirits to one’s own magical advantage. But in some instances ecstatic behaviour might be evaluated differently. The same persons might be viewed favourably by some as prophets, and condemned by others as false prophets or magicians. It seems likely that Jesus fell into this ambiguous category.

Spirit possession is clearly relevant to Jesus’ exorcisms, but its importance may well extend beyond these to his healings as well. For one thing, the Gospels occasionally appear to associate illnesses other than possession with evil spirits (Lk. 4:39; 11:14; 13:11; Mt. 9:32–33). This tendency is even more marked in some of the other texts we have discussed (e.g. the Genesis Apocryphon and the woes blamed on demons in much of the Pseudepigrapha). For another, many of the illnesses Jesus is reported to have healed (such as blindness, deafness and muteness) are of a type that in other cultures could be construed as the onset of a demonic attack. And for another, there is no clear indication that either Jesus or the Evangelists regarded the illnesses Jesus healed as being due to sorcery, the evil eye, or divine punishment. If Jesus conceived himself as a spirit-filled prophet overcoming demonic resistance to the reign of God through his healings and exorcisms, then it is plausible that he regarded other illnesses besides possession as being demonically caused too.

The obvious objection to this is that the Evangelists, particularly Mark, appear to distinguish sharply between exorcisms and healings (especially in redactional summaries) and that whereas Jesus healed by touch, he never touched demoniacs but al ways cast out demons by word of command. On the other hand, these distinctions may not be quite so sharp as supposed; not all of Jesus’ healings involve touch and many involve a word of command. Again, it may be that the distinction intended by Mark is not between demonic and non-demonic illnesses, but between different modes of demonic attack (possession of the sufferers’ minds or illnesses inflicted on their bodies).

Such arguments are hardly conclusive, however. It remains possible that Jesus regarded illnesses other than possession as due to impersonal causes, even if these were no more clearly defined than random misfortune and human frailty. As we have seen, this is suggested by applying Foster’s sickness taxonomy to Jesus’ ministry. The appearance of a naturalistic system may be admittedly due as much to the healing story form as to the beliefs surrounding that actual healings, yet the form is itself an indication of what people considered significant, so form-critical arguments hardly settle the question. Although it would be tempting to suggest that the question could be resolved by further research, the truth is probably that the Gospels simply do not preserve the information necessary to determine what type of disease aetiology Jesus worked with. The indications are nevertheless that he blamed illness on either demonic or impersonal causes rather than human malice or divine punishment.

The position is much clearer in the case of Jesus’ exorcisms, which clearly do presuppose a form of spirit possession. But even here it is necessary to exercise caution in applying anthropological insights. The precise circumstances of the demoniacs Jesus healed is not clear; the stories in the Gospels do not give us the kind of detailed case-history on which one could base either a psychiatric diagnosis or a general sociological theory about possession in first-century Palestine. What the exorcisms dob suggest is that Jesus himself may have been acting as a spirit-filled healer, able to cast out demons by the power of a more powerful spirit. This is clearly presupposed by the Beelzebul controversy, in which the point at issue is not whether Jesus is spirit-empowered but the nature of the empowering spirit: Beelzebul or the Spirit of God. It is also suggested by other indications in the Gospels. For example, in Mark the Spirit is said to enter into Jesus at his baptism (εἰς αὑτόν, 1:10) and then to drive him into the wilderness (ἐκβάλλει, 1:12). At Lk. 10:17–22 Jesus is said to have had a vision of Satan falling like lightning from heaven, and then to have ‘rejoiced in the Spirit’ and thanked his heavenly Father for special revelation; both are experiences suggesting some form of altered state of consciousness that could be related to spirit-possession.

In the Palestine of Jesus’ day, as Israel before, the only legitimate spirit-filled roles would be charismatic leader or inspired prophet. Otherwise, the spirit-possessed or spirit-controlling person would be seen as a demoniac or a magician. Which category one belonged to logically depended on the identity of the spirit in question, but in practice the identity of the spirit was likely to be determined by the category in which its human intermediary was placed. The category that comes closest to expressing Jesus’ self-understanding as divinely commissioned agent and messenger will almost certainly have been ‘prophet’. I shall return to the question of Jesus as prophet at the end of the concluding chapter.

In this chapter it has been possible to do no more than sketch the contribution social anthropology may be able to make to the topic. In doing so I have attempted to clarify the distinction between magic and miracle, I have indicated how social-scientific research into spirit possession may help to illuminate Jesus’ healings and exorcisms, and I have suggested how a model developed in connexion with a Mexican-American folk-saint may shed some light on Jesus’ role as folk-healer.

Chapter 14


General Observations

The evidence we have reviewed indicates that whereas most Second Temple Jews were by no means preoccupied with miracles, they did believe in them. There is little evidence of a dogmatic dispensationalism confining miracles to the biblical past, yet it is the great signs and wonders of old, particularly those associated with the exodus and conquest, that attracted the greatest attention. These, together with other great miracles of national deliverance (e.g. the slaughter of the Assyrian army by the angel of the Lord in the reign of Hezekiah), seem to have formed the basis for the liveliest hopes of signs and wonders in the future. Such hopes were not confined to the literate elite who produced the surviving documents, but were shared by the sign prophets and their popular following.

The healings and exorcisms of Jesus seem quite remote from this mainstream of Jewish expectations about signs and wonders. There are, however, a number of links. First, the healings attributed to Jesus bear some resemblance to those attributed to the great miracle-working prophets Elijah and Elisha. Secondly, there are occasional indications in Jewish literature of soteriological or even eschatological expectations of health, healing, and the defeat of demonic powers. Thirdly, the presence in the Jesus tradition of some sea and feeding miracles with at least superficial overtones of the exodus events (and an Elisha story) links Jesus more securely with the great signs and wonders of the past, and with Moses, who is both the Jewish figure through whom the greatest miracles were worked, and the figure with whom the great events of national deliverance and foundation were associated.

The remainder of this Conclusion will unpack these observations under the headings ‘Healing and Exorcism’, ‘Other Miracles’, and ‘Jesus as Prophet—and More than a Prophet’.

Healing and Exorcism

The Jewish texts we have examined show comparatively little interest in miracles of healing, neither is there much evidence for other Jewish healers around the time of Jesus. Josephus records few miracles of healing; Philo records none. The Genesis Apocryphon narrates a healing/exorcism of sorts, as does the book of Tobit, and there are one or two other places in the Apocrypha and Pseudepigrapha where healing is mentioned or a biblical healing recalled. On the whole, though, other types of miracle predominate, notably punishment miracles, deliverance miracles and provision miracles. Ḥanina ben Dosa may have been an effective intercessor, but there is little to suggest that he engaged in a ministry of healing in any way comparable to Jesus’, and nothing to suggest that he was one of a class of charismatic healers operating at the time. The so-called ‘sign prophets’ mentioned by Josephus often promised miraculous signs, but none of them is reported either to have healed or to have promised healing.

There is some indication that Jewish exorcists were active in the first century, although the evidence is insufficient to be able to say that they were particularly common. Both in his description of the Baaras root and in his account of Eleazar, Josephus creates the impression that exorcism was closer to a magic art that an irruption of divine power. The incident with the sons of Sceva in Acts 19 also portrays Jewish exorcists as magicians (whether or not Sceva was himself a Jew), although this may be due to Lukan polemic. On the other hand, both Tobit and the Genesis Apocryphon describe demons being chased off with divine backing, although in the latter text the evil spirit seems to have been under God’s control all along. Overall, it is in the Enochic and Qumran literature that demons and apotropaic counter-measures (not necessarily exorcisms) play the largest role.

This leaves Jesus as unique in the surviving Jewish literature of his time in being portrayed as performing a large number of healings and exorcisms. Kahl is correct that he is virtually unique in Jewish literature in being portrayed as an immanent BNP in these acts of power, with two provisos. First, some Jewish texts can appear to make human figures act as immanent BNPs where the context suggests that they are really only mediators of God’s numinous power. Secondly, whatever may be true of individual miracles stories, the Gospels show some tendency to indicate that it is God’s power at work in Jesus, not merely his own (e.g. Mt. 12:28//Lk. 11:20; Lk. 5:17). If Jesus is a BNP, it is because he is a bearer of God’s Spirit, which is the source of Jesus’ power. Indeed, if God were not in some sense behind Jesus’ acts of power, they would not count as miracles.

To be sure, the literate elite who produced most of the surviving texts may not have been typical of Jews as a whole, and miraculous healing may have been of much more interest to Galilean peasants than it appears to have been to either Josephus or Philo. Philo (in common with Ben Sira) gives the impression that the physicians he knew were competent and beneficial; it is unlikely that this level of professional health care was available to peasants in Galilee. Cross-cultural comparison would suggest that the bulk of primary health care in Galilee would be in the popular sector, but that it is likely that a number of folk-healers would also have been available. Unfortunately, we have no means of knowing how Galilean folk-healers operated, but in respect of his healing activities Jesus would certainly fall into this category.

Clearly, however, Jesus was no ordinary folk-healer. Romano’s study of the Mexican-American folk healer Pedrito Jaramillo shows how wholehearted dedication to the healing role can propel the healer up the healing hierarchy to folk-saint, and something of the same may have happened to Jesus. Of course there are significant differences, but it is noteworthy how a conviction of divine calling, a renunciation of normal family and economic ties, a peripatetic ministry with striking methods of healing, and a conservative charisma activating a latent ideal may have been common features contributing to the growing fame of both men. Unlike Pedrito, however, Jesus also seems to have been remarkable as an exorcist.

There are a few Jewish texts that express an eschatological hope connected with healing, the most notable being the Messianic Apocalypse (4Q521) from Qumran and the promise of final health and longevity at Jub. 23:23–31. In the first instance these texts probably intend healing as a metaphor for national restoration rather than envisaging a series of individual healing miracles. Nevertheless, the use of this type of metaphor might well have allowed such miracles to be employed as acted signs or parables of national restoration, the coming of God’s kingdom, or some other form of eschatological fulfilment. The hope that the demonic forces of evil would be overthrown at the eschaton is fairly well attested, although Sanders is correct to say that nowhere is it suggested that this overthrow would be achieved through exorcisms. Yet this idea could have come about through the mingling of streams that otherwise seem separate in the surviving literature: stories of exorcisms such as that performed by Eleazar and narrated by Josephus; popular demonology perhaps reflected in some of the Dead Sea Scrolls and presumably experienced as various forms of worrying spirit possession, and the demonological myths of the Enochic literature together with apocalyptic hopes for the final defeat of evil powers. A partial merging of these streams already seems to have taken place in Jubilees (which begins to assimilate the roles of evil spirits and fallen angels) and Tobit (where there is an exorcism of a demon, Asmodeus, who seems to have some points of contact with the Enochic fallen angels).

Again, both in some of the Qumran texts and even more in Philo healing was used as a metaphor for spiritual healing, that is, for healing from sin for a life of virtue. This indicates that physical healing may have been available as an acted sign, not only of God’s graciousness towards his people, but of the need for turning to God for inner as well as outward healing (Mk 2:5?). In any case, these admittedly scattered and fragmentary hints are sufficient to suggest that there were some hopes connected with healing that Jesus could tap into. Not only would healing and exorcism be immensely popular among people without access to professional health care, but there were hopes latent within the Judaisms of the day that could be used to give such acts a deeper meaning.

Theissen is thus broadly correct in suggesting that Jesus’ uniqueness lay in seeing eschatological salvation realized in individual acts of healing and exorcism. For one thing, eschatological or salvation-historical miracles are generally envisaged as being on a far larger scale than the healing of individuals. For another, eschatological or salvation-historical miracles are virtually never envisaged as miracles of healing (or exorcism), though this needs to be seen against a relative lack of interest in healing miracles and exorcisms in general.

How, then, does Jesus come to attach eschatological significance to healing and exorcism? The explanation may lie along the following lines. First, healing and exorcism were, perhaps, the only type of miracles available to Jesus. Secondly, he was nevertheless an extraordinarily gifted healer and exorcist who, even more than Pedrito Jaramillo, excelled among the folk-healers of his time. Thirdly, he understood this gift as an even greater empowerment by God’s spirit than that of the great prophets of old. This experience of divine empowerment (coupled with a special sense of a close relationship with God) prompted Jesus to reinterpret, combine and transform Jewish traditions. He then understood his individual acts of healing and exorcism as prophetic signs, not only of the imminence of the Kingdom of God, but of its nature.

Other Miracles

Second Temple literature does show more interest in the more spectacular types of miracle, particularly those where God intervenes to ensure victory for Israel, or to inflict a punishment on its enemies or the wicked within, or to deliver the righteous from peril. The spectacular miracles associated with the exodus tradition are often recalled, and are narrated or alluded to in nearly all the types of literature we have examined. These and other Old Testament miracle stories are used by Josephus to illustrate God’s providence working in striking ways. He also uses them to show how God authenticated his prophets. In Philo they appear more as proofs of divine activity, but also serve to underline the roles of Moses as king, prophet, priest and law-giver. Elsewhere they are frequently appealed to in the hope that God will intervene in like manner again on behalf of his beleaguered people.

Compared with these great miracles, the Gospel ‘nature’ miracles appear almost tame. In particular, the Gospel feeding and sea miracles that are often taken as allusions to the exodus events lack the spectacular characteristics found in allusions to the exodus miracles in other Jewish literature. This is not to deny that these stories are given exodus allusions in the Gospel narratives as we have them, but it is to question whether they can have been created for this purpose.

To begin with, the Gospel feeding stories are unlike both current expectations of eschatological plenty or references to the manna miracle. Several thousand people eat and are satisfied, but they have a plain meal of bread and fish for sustenance, not a splendid feast. This lacks the superabundant grandeur of vines bearing vast clusters of giant grapes, or beached leviathans providing huge quantities of meat. It also lacks an essential feature of virtually every narration of the manna story, namely that the manna came down from heaven. In the Gospels, the crowds are fed from purely earthly loaves, not from anything that falls from the sky. Again, the only miracle is that so many people are satisfied from such meagre resources, there is no indication, as in the manna story in Exodus and its imitators, that the food was in any way exotic or unusual.

The one Old Testament miracle story that the Gospel feeding stories do plainly resemble is that of Elisha feeding a hundred men from twenty loaves of barley. Although there is little verbal similarity between any Gospel feeding account and the lxx of 4 Kgdms 4:42–44 (beyond the phrase ἄρτους κριθίνους in Jn 6:9 and 4 Kgdms 4:42), an analysis of the structure of the two stories would show a great deal in common. Strangely, this Elisha story is not referred to anywhere else in surviving Second Temple literature, even in places where one might expect to find it, such as Jewish Antiquities or Lives of the Prophets This makes it unlikely that the Gospel feeding stories were pure invention. Had the Evangelists or the tradition behind them wished to invent a provision miracle for Jesus based on a famous Old Testament precedent, they would have been far more likely to have chosen the renowned story of Moses and the manna in the wilderness than the relatively neglected story, ignored by both Josephus and the Lives of the Prophets, of Elisha and the barley loaves.

Similar observations may be made about the two sea miracles attributed to Jesus. For one thing, they do not have the standard features found in every other narration of the Red Sea crossing. The Gospels’ redaction makes them suggest the Red Sea in a number of ways, notably the conjunction with the feeding story in Mark 6 and John 6. Mk 4:35–5:20 may also hint at the drowning of pursuing enemies by drowning the possessed pigs and demonizing the wind and waves. But these suggestions arise from the secondary arrangement of individual pericopae, not from the pericopae in themselves. Neither of Jesus’ sea miracles involves flight from enemies, parting the waves, crossing on dry land, or the drowning of hostile pursuers, all of which are features emphasized in contemporary accounts of the Red Sea events or miracles based upon it. Had the Gospel sea stories been invented to recall the exodus, they would surely have done so far more closely. As it is there is only a generic resemblance in Yahweh’s control of the wind and water, but no specific resemblance to the events at the Red Sea.

These stories are nevertheless thoroughly Jewish. They depend for their significance on the Old Testament idea that it is Yahweh who rules the waves and tames the waters of chaos. The stilling of the storm echoes both Jon. 1:4–15 and Ps. 107:23–31. The walking on the sea has some affinity with both these passages and also with Job 9:8, Hab. 3:15; Isa. 51:9–10; and Ps. 77:19 (among others). The strange notice in Mk 6:48 that Jesus meant to pass the disciples by is likewise most probably an echo of Yahweh’s passing by Moses (Exod. 33:19, 22) and Elijah (1 Kgs 19:11).

These stories thus act in two contrary ways. On the one hand, they are comparatively restrained. They are not so spectacular as to invite instant disbelief. One can at least picture Jesus and his disciples in a boat during a storm, or Jesus distributing loaves to a crowd, whereas a shower of manna from the sky or the waters of the Sea of Galilee piling up to allow Jesus and the Twelve to walk across the dry lake bed would read far more like legendary echoes of the Old Testament. This is not to deny that the feeding and water miracles are highly problematic from the viewpoint of modern science (and, no doubt, from that of ancient common sense), but they seem to have been told with some regard to plausibility. Some historical constraints appear to be operating here; these stories are not simply pure inventions to make theological points.

On the other hand, they do make theological points. For one thing, they link Jesus to the traditions of the great signs and wonders of the past more clearly than do his healings and exorcisms. Yet by not imitating the Red Sea stories the Gospel sea stories avoid casting Jesus in the role of Moses. Moses was probably the most revered human figure in Judaism, and there are clear Mosaic elements in Matthew’s portrayal of Jesus. But in these sea stories Jesus is portrayed as filling the role, not of another Moses, but of Yahweh. In the feeding stories, on the other hand, Jesus is made more like Elisha than Moses. The Fourth Gospel, which picks up the potential exodus echoes in this story most explicitly, uses them to make a polemical contrast between Jesus and Moses.

What seems to be going on here is a desire to place a strong theological overlay on Jesus as he was known to have been, rather than some Jesus of pure fantasy. If the Jesus of the sea and feeding miracles is seen through the prisms of both the Easter faith and the Old Testament, it is nevertheless the real Jesus that is seen. Behind the feeding stories most probably lies the genuine reminiscence that Jesus attracted large crowds and on one or more occasions somehow arranged for them to fed. Behind the stilling of the storm there may well lie the reminiscence of one or more occasions in which Jesus and his disciples encountered a sudden storm on the lake.

The theologizing evident in these stories is also relevant to Kahl’s thesis. Kahl’s study covered only healing stories, but he mentioned that his study should be extended to nature miracles. The present study has found that in Jewish nature miracles the human agent is nearly always portrayed as PNP or MNP but not BNP. The BNP is nearly always God, except in these Gospel stories about Jesus. Since, however, both the feeding stories and the sea miracles make best sense against an Old Testament background, it is implausible to appeal to pagan influence in the tendency to make Jesus a BNP in them.

Jesus as Prophet—and More than a Prophet

This survey of Jewish literature has shown that although various different types of human figure may be associated with the working of miracles, prophets form by far the largest category. This is especially so if Moses is counted among the prophets, since in both Philo and Josephus, as in many other texts, a large proportion of the miraculous activity is associated with Moses. We have also seen that the characteristics of a spirit-filled prophet match the traditions about Jesus at several points. That is to say that his power to heal and exorcise would almost certainly have been seen as due to some kind of spirit. If it was an evil spirit then he would be a sorcerer and a false prophet. If it was the Spirit of Yahweh, the Holy Spirit, then he would have been a true prophet like the prophets of old, at least some of whom must have been similarly inspired. One cannot make a decision between calling him a true prophet or a false prophet without at the same time making a theological commitment. If one wishes to be strictly neutral, one must use the word ‘prophet’ as denoting a social type without regard to the presence or absence of divine legitimation.

Further indirect support for this classification comes from examining the other miracle-working figures at the time. There is nothing to suggest the existence of a class of charismatic holy men to which Jesus might have belonged. He does not closely resemble the few other Jewish exorcists we know about. On the other hand, although there are significant differences between Jesus and the ‘sign prophets’, the existence of the latter does show that there were people at the time willing to entertain prophetic claims and to associate them with the promise of miraculous signs. These sign-prophets also attracted large crowds of followers. It is thus plausible to suppose that someone who worked miraculous healing, announced a message from God (the coming of his kingdom) and attracted a large following would be seen as a prophet in that time and culture, and would intend to be seen thus.

Yet Jesus is quite distinctive as a prophet. Unlike the sign-prophets who came after him he healed and exorcized, but seems not to have promised a particular spectacular sign. Conversely, healing and exorcizing were not that essential to the prophetic role. Few of Israel’s classic prophets were healers. In the story about Isaiah healing Hezekiah’s sickness, the miracle performed at Isaiah’s behest is the sign of the reversing shadow, not specifically the healing. Moses once or twice halts or reverses illness inflicted as a punishment, but he is not represented as a healer. Only with Elijah and Elisha does healing appear to enter the prophet’s role, and even then not very often.

The miracles narrated of Jesus make him resemble Elisha more closely than either Moses or any other figure from the Old Testament, although the resemblance is not that great. In performing multiple miracles, mostly in what had once been the Northern Kingdom of Israel, over an extended period he was also more like Elisha than either the sign prophets or any other of his Jewish contemporaries known to us. Yet he hardly played an Elisha-like role in the affairs of Israel.

More importantly, Jesus differs from all other prophetic figures known about in Judaism (with the possible exception of Artapanus’s Moses) in performing his miracles as a BNP. It could, of course, be argued that this is simply a feature of the Gospel depiction of him, coloured by Easter faith, and not at all an accurate observation about the historical Jesus. But two considerations point the other way. The first is the consistency in the portrayal of Jesus as BNP rather than MNP or PNP throughout the tradition, despite the rarity of such a portrayal of a human figure elsewhere in Judaism. The second is the parallel with the way Jesus spoke in his own name, rather than employing the traditional prophetic messenger formula ‘Thus says Yahweh.’ Thus, whereas it is perfectly true that in the Judaism of his time, no amount of miracle-working would have been accepted as proof of divinity, the miracles worked by Jesus may contribute as much to an implicit Christology as the authority of his teaching, as the earliest Evangelist realized long ago (Mk 1:22, 27).

This study has been confined to the Jewish context of Jesus’ miracles. It has demonstrated, inter alia, how both the Gospel presentations of Jesus’ miracles and the miracles of the historical Jesus would appear distinctive within the Judaism of his time while making sense in a Jewish context. This by no means exhausts the task of understanding Jesus’ miracles, however. Clearly, one would need to undertake a similar study of roughly contemporary pagan miracle stories in order to ascertain how far what appears distinctive about the miracles of Jesus in a Jewish context may be due to the Graeco-Roman cultural influences on the Gospel tradition. Such a study would need to take into account the uses to which the pagan stories were being put as much as their structure and motifs (as in Kahl’s study of healing stories). One could also pursue social-scientific insights into medical anthropology, magic and spirit-phenomena in far more depth than it has been possible to do here. And finally, one would need to return to the New Testament and other early Christian texts for a fresh analysis of the relevant material. In this conclusion it has been possible only to indicate some of the lines such an analysis might pursue. The study as a whole has hopefully served to clarify the role of miracle in Second Temple Judaism generally, and in such a way that the Jewish context of Jesus’ miracles comes into sharper focus.


Africa, Thomas, ‘Worms and the Death of Kings: A Cautionary Note on Disease and History’, Classical Antiquity 1 (1982), pp. 1–17.

Alexander, P.S., ‘Incantations and Books of Magic’, in Schürer (ed.), The History of the Jewish People in the Age of Jesus Christ, III. 1, pp. 342–79.

———’ “Wrestling against Wickedness in High Places”: Magic in the Worldview of the Qumran Community’, in Craig A. Evans and Stanley E. Porter (eds.), The Scrolls and the Scriptures: Qumran Fifty Years After (JSPSup, 26; Sheffield: Sheffield Academic Press, 1997), pp. 318–37.

Anderson, Hugh, The Gospel of Mark (NCB; London: Marshall, Morgan & Scott, softback edn, 1981).

Attridge, Harold W., The Interpretation of Biblical History in the Antiquitates Judaicae of Flavius Josephus (Missoula, MT: Scholars Press, 1976).

———’Historiography’, in Michael E. Stone (ed.), Jewish Writings of the Second Temple Period (Philadelphia: Fortress Press, 1984), pp. 157–84.

Aune, David E., ‘Magic in Early Christianity’, ANRW, II.23.2 (1980), pp. 1507–57.

Bammel, Ernst, ‘ “John did no miracle”: John 10:41’, in C.F.D. Moule (ed.), Miracles: Cambridge Studies in their Philosophy and History (London: Mowbrays, 1965), pp. 181–202.

Barclay, J.M.G., Jews in the Mediterranean Diaspora from Alexander to Trajan (323 bce–117 ce) (Edinburgh: T. & T. Clark, 1996).

Barnett, P.W., ‘The Jewish Sign Prophets, ad 40–70: Their Intentions and Origin’, NTS 27 (1981), pp. 679–97.

Baroja, Julio Caro, ‘Magic and Religion in the Classical World’, in Marwick (ed.), Witchcraft and Sorcery, pp. 73–80.

Bauckham, Richard, ‘The Liber Antiquitatum Biblicarum of Pseudo-Philo and the Gospels as “Midrash” ‘, in R.T. France and David Wenham (eds.), Gospel Perspectives, III. Studies in Midrash and Historiography (Sheffield: JSOT Press, 1983), pp. 33–76.

———The Fate of the Dead: Studies on the Jewish and Christian Apocalypses (NovTSup, 93; Leiden: E.J. Brill, 1998).

Baumgarten, Joseph M., ‘The 4Q Zadokite Fragments on Skin Disease’, JJS 41 (1990), pp. 153–65.

Beasley-Murray, George R., Jesus and the Kingdom of God (Exeter: Paternoster Press, 1986).

Beentjes, Pancratius C. (ed.), The Book of Ben Sira in Modern Research (BZAW, 255; Berlin; New York: W. de Gruyter, 1997).

Begg, Christopher, Josephus’ Account of the Early Divided Monarchy (BETL, 108; Leuven: Leuven University Press, 1993).

Berchman, Robert M., ‘Arcana Mundi: Prophecy and Divination in the Vita Mosis of Philo of Alexandria’, in David J. Lull (ed.), SBL 1988 Seminar Papers (SBLSP; Atlanta: Scholars Press, 1988), pp. 385–423.

Berman, Dennis, ‘Hasidim in Rabbinic Traditions’, in Paul J. Achtemeier (ed.), SBL 1979 Seminar Papers (SBLSP; Missoula, MT: Scholars Press, 1979), II, pp. 15–33.

Betz, Otto, ‘The Concept of the So-Called “Divine Man” in Mark’s Christology’, in D.E. Aune (ed.), Studies in New Testament and Early Christian Literature (NovTSup, 33; Leiden: E.J.Brill, 1972), pp. 229–40.

———’Das Problem des Wunders bei Flavius Josephus im Vergleich zum Wunderproblem bei den Rabbinen und im Johannesevangelium’, in Otto Betz, Klaus Haacker and Martin Hengel (eds.), Josephus-Studien: Untersuchung zu Josephus, d. antiken Judentum u. d. Neuen Testament. Otto Michelz. 70 Geburtstag gewidmet (Göttingen: Vandenhoeck & Ruprecht, 1974), pp. 23–44.

———’Miracles in the Writings of Flavius Josephus’, in Louis H. Feldman and Gohai Hata (eds.), Josephus, Judaism and Christianity (Leiden: E.J. Brill, 1987), pp. 212–35.

Betz, Otto, and Rainer Riesner, Jesus, Qumran and the Vatican (London: SCM Press, 1994).

Bilde, Per, Flavius Josephus between Jerusalem and Rome: His Life, his Works and their Importance (Sheffield: JSOT Press, 1988).

Blackburn, Barry L., ‘ “Miracle Working Theioi Andres” in Hellenism (and Hellenistic Judaism)’, in David Wenham and Craig Blomberg (eds.), Gospel Perspectives. VI. The Miracles of Jesus (Sheffield: JSOT Press, 1986), pp. 185–218.

———Theios Anēr and the Markan Miracle Traditions: A Critique of the Theios Anēr Concept as an Interpretative Background of the Miracle Traditions Used by Mark (WUNT, 2; Tübingen: J.C.B. Mohr [Paul Siebeck], 1991).

Boccaccini, Gabriele, Beyond the Essene Hypothesis: The Parting of the Ways between Qumran and Enochic Judaism (Grand Rapids: Eerdmans, 1998).

Boddy, Janice, ‘Spirit Possession Revisited: Beyond Instrumentality’, Annual Review of Anthropology 23 (1994), pp. 407–34.

Boers, Hendrikus, Who Was Jesus? The Historical Jesus and the Synoptic Gospels (San Francisco: Harper & Row, 1989).

Bokser, Baruch M., ‘Wonder-Working and the Rabbinic Tradition: The Case of Hanina ben Dosa’, JSJ 16 (1985), pp. 42–92.

Booth, R.P., Contrasts: Gospel Evidence and Christian Beliefs (Bognor Regis: Paget Press, 1990).

Borg, Marcus J., Jesus: A New Vision. Spirit, Culture, and the Life of Discipleship (San Francisco: Harper & Row, 1987).

Bow, Beverly, and George W.E. Nickelsburg, ‘Patriarchy with a Twist: Men and Women in Tobit’, in Amy-Jill Levine (ed.), Women Like This’: New Perspectives on Jewish Women in the Greco-Roman World (Early Judaism and its Literature, 1; Atlanta: Scholars Press, 1991), pp. 127–43.

Braun, Martin, History and Romance in Graeco-Oriental Literature (Oxford: Basil Blackwell, 1938).

Brockington, L.H., A Critical Introduction to the Apocrypha (London: Gerald Duckworth, 1961).

Brown, Colin, Miracles and the Critical Mind (Exeter: Paternoster Press, 1984).

———That You may Believe: Miracles and Faith Then and Now (Exeter: Paternoster Press, 1985). Brown, P., ‘The Rise and Function of the Holy Man in Late Antiquity’, JRS 61 (1971), pp. 80–101.

Bruce, F.F., The Acts of the Apostles: The Greek Text with Introduction and Commentary (London: Tyndale Press, 1951).

———’Jesus and the Gospels in the Light of the Scrolls’, in Matthew Black (ed.), The Scrolls and Christianity (London: SPCK, 1969), pp. 70–82.

Büchler, Adolf, Types of Jewish-Palestinian Piety from 70 B.C.E. to 70 C.E: The Ancient Pious Men (Jews’ College Publications, 8; London: Jews’ College, 1922).

Bultmann, Rudolf, The History of the Synoptic Tradition (trans. John Marsh; Oxford: Basil Blackwell, 1963).